Contractor’s Financial Estimation

based on Owner Payment Histories

Hanh Tran David G. Carmichael
School of Civil and Environmental School of Civil and Environmental
Engineering, The University of New South Engineering, The University of New South
Wales, Sydney, 2052 NSW Australia Wales, Sydney, 2052 NSW Australia
D.Carmichael@unsw.edu.au

a contractor ’s financial viabilit y is affected by late and incomplete pay -

DOI 10.5592/otmcj.2012.2.4
Research paper ments from the owner . Late and incomplete payments lead to cash flow
uncertainty, additional bank interest, and delays in paying creditors
such as suppliers and subcontractors, and may lead to decreased proj-
ect performance, and possible additional time and cost due to disputes.
The paper presents a method for cash flow and present value analysis
under uncertainty based on an owner’s payment history or estimated
payment characteristics. The paper generalises existing modelling of
uncertainty associated with late and incomplete owner payments to a
range of claim types by the contractor, and different owner types. Aging
contractor claims are analysed for claims submitted on a regular basis
for amounts which may vary depending on project phasing. For each of
the pre-identified typical owner payment practices, the estimated paid
proportions of claims and the steady state distribution of payments in dif-
ferent age categories are established. A present value analysis assesses
project viability from the contractor’s viewpoint. Actual project data are
used to confirm the validity of the method. The intent of the paper is to
assist contractors establish suitable allowances in their tender pricing,
Keywords
to choose a suitable claim/payment schedule and/or to adopt suitable
Cash flow, Markov chains,
administration practices to optimise cash flow. The paper gives a sum-
contractor payment,
mary approach for contractors, providing them with a practical tool in
owner classification
cash flow planning, control and risk management.

hanh tran · da vi d g . ca r m i cha e l · c o n t r a c t o r ’ s f i n a n c i a l e s t i m a t i o n b a s e d o n o w n e r  … · pp 481 - 489 481

 INTRODUCTION (Tran and Carmichael, 2013), the esti- also serve as a reference tool for project
Cash flow forecasting and cash flow man- mated paid proportions of claims and owners to enhance their relationship
agement are essential but difficult as- the steady state distribution of pay- with contractors. The paper provides
pects of a contractor’s practices; they are ments in different age categories are a practical tool for cash flow planning
central to the wellbeing of a contractor. established. Real project data are used and management; it is a contribution to
Forecasting is also important as a means to confirm the validity of the method. contractor financial planning and risk
to obtain loans, because banks prefer to The aim of the paper is to assist con- management.
lend money to companies that can pres- tractors in establishing a detail cash The paper starts by reviewing related
ent periodic cash flow forecasts (Navon, flow forecast which takes into account studies about claim-payment modelling
1995). However a contractor’s cash flow cash inflow uncertainties due to late and and cash flow estimation and then sum-
is subject to many uncertainties, some of incomplete owner payment behaviour, marises some key results from the litera-
which result from owner payment prac- and cash outflow. As a follow-on, con- ture. The existing literature is modified to
tices. An owner which fully complies with tractors are able to establish suitable incorporate claims that change with proj-
payment terms outlined in the conditions allowances in their tender pricing or to ect phasing. Case study data are used to
of contract makes cash flow management choose a suitable claim/payment sched- demonstrate the cash flow and present
much easier, while an owner which re- ule to optimise cash flow. The paper’s value calculations, taking into account
sponds irregularly and incompletely to method can be used to address risks alternatives in payment time lags to sub-
a contractor’s claims may drive the con- associated with negative cash flow, ad- contractors and mark-up in claims.
tractor’s cash flow to deviate far from ditional bank interest, and disputes,
what had been planned. An understand- leading to more effective cash manage- Background Literature
ing of an owner’s payment practices is, ment by the contractor. Uncertainties in payments leading to
therefore, very useful for a contractor’s Although in some countries there cash flow difficulties have been high-
cash flow planning purposes. exists legislation to protect contractors lighted as a cause of business failures
The paper presents a method for from late and incomplete payments and and escalating disputes (Carmichael,
cash flow and present value (equiva- insolvency of the payer, payment ar- 2002; Carmichael and Balatbat, 2010).
lently present worth) analysis under un- rears are still very common (Wu et al., Some research has attempted to assist
certainty based on an owner’s payment 2011; Brand and Uher, 2010). Owner- in mitigating construction uncertainties
history or estimated payment charac- caused delays and incompleteness in associated with claims and disputes.
teristics. Extending from the original payments have been shown to have a Examples include predicting contrac-
work of Carmichael and Balatbat (2010), large influence on a contractor’s cash tor failure (Russell and Zhai, 1996),
the method gives the change in claim flow and financial viability (Carmichael, evaluating and investing in construc-
payments in weeks/months following 2000, 2002; Carmichael and Balatbat, tion projects under uncertainty (Ho and
claim lodgement. Payments of indi- 2010). Cost and time associated with Liu, 2003), and developing an integrated
vidual claims are accumulated and su- disputes may also place a large burden method for project risk management
perimposed on the planned cash out- on contractors. An example given by El- from the owner’s point of view (del Cano
flows throughout the project, so that Adaway and Kandil (2009) emphasises and de la Cruz, 2002).
a detailed cash flow diagram can be the severe losses to a contractor when Cash flow forecasting is about the
obtained. A present value analysis is it had to wait for a 3-year arbitration distribution of income and expenditure
performed to assess project viability to run its course before recovering the as a function of time (Navon, 1995). It is
from the contractor’s viewpoint. majority of its claim. noted that the majority of existing pub-
Payment time lag to creditors such The method presented in this paper lications about cash flow forecasting
as subcontractors, owner type (repre- can be combined with the Carmichael- focus on expenditure, which is taken
sented by different payment profiles) Balatbat Markov chain formulation of from the project schedule. For exam-
and claim mark-up are analysis vari- owner payments and the classification ple, Navon (1995) introduces a resource-
ables. Claims are allowed to change of owner payment behaviour (Tran et based cash flow estimation, Kenley and
in line with project phasing and typi- al., 2011; Tran and Carmichael, 2012a,b, Wilson (1986, 1989) model project net
cal project S curve behaviour. Aging 2013) to form a complete cash flow anal- cash flow as a logit-transformation of
contractor claims are assumed to be ysis tool. While primarily intended for percentages of project time and cost,
submitted on a regular basis; claim contractors, the method can also be Chen and Chen (2000) integrate a cost
amounts may vary depending on proj- used by subcontractors, suppliers and database and billing activity payments
ect phasing. For each of the pre-identi- consultants when they deal with others of subcontractors into the cash flow es-
fied typical owner payment practices higher in the contractual chain. It may timate, and Kaka and Price (1993) sim-

482 o rga n i za t i o n , te ch n ol o g y a n d ma na ge m e n t i n co nst r u c t i o n · an international journal · 4(2)2012

plify the standard cost-commitment of payment by a certain date and the Extension of the Carmichael-
curve to enable contractors to perform average time to payment. The present Balatbat Formulation
cash flow estimates at the pre-tender- analysis follows this line of thinking but for Calculating Changes
ing stage more readily. Blyth and Kaka allowing for different claim submission in Payments
(2006), Hwee and Tiong (2002), and Ma- schedules that reflect project phasing. The Carmichael-Balatbat formulation
vrostas et al. (2005), among others, use can be used to estimate the change in
a project’s S curve as a guide for es- Background Theory amounts in the transient states follow-
timating cash outflow; an underlying Carmichael and Balatbat (2010) model ing a claim submission. Consider a claim
assumption in these cash flow forecast contractor payments by owners using of value c1. A claim is equivalent to an
models is that payments occur as antici- Markov chains in the following sum- amount (here c1) entering state 0, with
pated pre-project. mary way. zero amounts in the other states 1, 2,
Some studies that discuss changes Let period i = 0 be the time that the …, n. These other states only take val-
in cash inflow are Park et al. (2005), claim is made by the contractor; then ues when transitions between states oc-
Chen et al. (2005), Kaka and Price (1991) periods i = 1, 2, 3, ... represent months/ cur. Accordingly define 1 × n row vector
and Kaka (1996). The model by Park et weeks beyond that time. Let the (tran- C1 = [c1, 0, 0, ..., 0] as the vector of new
al. (2005) allows contractors to incor- sient) states be the amount outstand- state additions. Over one time period, the
porate the time lag between expendi- ing to the contractor beyond period i. amounts in the transient states change
ture and payment of a related cost item. The states reflect the aging amount be- to C1Q, over two time periods to C1Q2 and
Chen et al. (2005) recommend the in- lieved by the contractor to be owed on so on.
clusion of more detailed payment con- the project. Two additional (absorbing) In the following time periods i = 2,
ditions, and differential payment lags states n’ and n are also introduced. n’ 3, ..., (here a month, week, ...), allow
and frequency in order to increase the is the ‘Paid’ state and n is the ‘To be re- claims respectively of c2, c3, .... And so,
accuracy of cost-schedule integrated solved’ state. As noted, states 0, 1, 2, …, using equivalent notation as above, the
cash flow forecasting techniques. Kaka n-1 are referred to as transient states, amount in each transient state contrib-
(1996) mentions payment delays and while states n and n’ are referred to as uted by the latest claim (after 0 time pe-
retention in cash flow calculations, as- absorbing states. riods) is Ci, contributed by the previous
suming that delay is minimal and the Transition probabilities between claim (after 1 time period) is Ci-1Q, con-
work in progress and the value of prog- periods i and i+1 are calculated from, tributed by the claim before that (after
ress claims are equal. 2 time periods) is Ci-2Q2, and so on. The
Doubtful accounts in retail busi- cumulative amount in each transient
nesses are modelled as Markov chains

 state contributed by all claims is Ci +
by Cyert et al. (1962) to estimate col- Ci-1Q + Ci-2Q2 + ...
lectibles and the probable time to col- For claims of constant amounts C1 =
lection. The estimates of collectibles C2 = ... = Ci = C, the steady state amount
are then calculated for the case where j, k = 0, 1, 2, ..., n, n’   (1) in each transient state is
monthly inputs of claims vary cyclically
as occurs in retail businesses. There Here α is the amount in state k that is C + CQ + CQ2 + CQ 3 + ... =
are several modifications to and com- transferred from state j between peri- C(1 + Q + Q2 + Q 3 + ...) = CN   (2)
ments on the original contribution of ods i and i+1.
Cyert et al. (1962), including Corcoran pjk , for j, k = 0, 1, 2, …, n, n’, com- CN is a 1 x n vector.
(1978), van Kuelen et al. (1981), Bark- prise the elements of an transition ma-
man (1977), Wort and Zumwalt (1985), trix P which is partitioned to give Q (n A similar argument can be used for
Kallberg and Saunders (1983) and Fry- x n) and R (n x 2) matrices. R applies the absorbing states. Of any new claim,
dman et al. (1985). to transitions from transient states to CR will be absorbed, of the preceding
Carmichael and Balatbat (2010) use absorbing states, while Q applies to time period claim CQR will be absorbed,
Markov chains to model late and incom- transitions between transient states. of the time period CQ2R claim before
plete owner payments. States are de- A fundamental matrix N = (1–Q) -1, and that will be absorbed, and so on. That
fined as the period of time by which pay- a matrix NR are then computed. The is, the steady state amount absorbed is
ment is overdue. Transition probabilities first column of NR gives the probabili- CR + CQR + CQ2R + CQ 3R + ... =
are estimated based on summaries of ties of amounts being paid. The second C(1 + Q + Q2 + Q 3 + ...)R = CNR   (3)
total project outstanding amounts over column of NR gives the probabilities of
time. The analysis gives probabilities amounts needing resolution. CNR is a 1 x 2 vector.

hanh tran · da vi d g . ca r m i cha e l · c o n t r a c t o r ’ s f i n a n c i a l e s t i m a t i o n b a s e d o n o w n e r  … · pp 481 - 489 483

 Thus, the claim submission schedule claim of $143.9K in the second month … of $0, $63.6K, $108.4K and so on.
of the contractor can be converted to an gives rise to payments of $0, $96.3K, Figure 1 plots shows the claim-
estimate of future cash inflow, which can $18.3K and $0 in subsequent months. payment relationship for the project,
be combined with planned cash outflow. And continuing, strings of payments of in which the payments are calculated
Actual claim submission schedules can claims in the third, fourth, etc. months using the above extension of the Car-
be used. Alternatively, constant claims can be calculated. Summing the pay- michael-Balatbat formulation.
within project phases may be assumed. ment contributions of each claim leads The total claimed value was ap-
Below, a case study project is used to to the total payments in months 1, 2, 3, proximately $1,134.2K, while the to-
demonstrate the method.

Outstanding amount ($K) at

Case Example A Total claimed amount ($K)
Consider the claims and payment data 1 month 2 months 3 months 4 months
for the construction of noise-reduction 1,134 1,134 375 231 231
walls along a metropolitan railway
line. The project contains 12 progress Table 1 Outstanding claimed amounts at months following claim lodgement –
claims totalling approximately $1.2M. case example A; n = 4
The project duration was approximately
12 months. Two progress claims were 200.0
not paid and the reasons given were
150.0
that the work had not been completed,
or insufficient detail was submitted in 100.0
the progress claim. Table 1 shows the
50.0
summary of the outstanding project
Amount ($K)

money against the number of months 0.0

after claims lodgement.
-50.0 3 5 11 12 13 14 15
Based on the payment profile in Ta-
ble 1, the matrices Q and R can be as- -100.0
sembled according to Carmichael and
-150.0
Balatbat (2010),
-200.0

-250.0 Time (month)

  Figure 1: Claim-payment relationship - case example A

10 140

and 9
120
8
Cummulative amount ($M)

7 100
Claim amount ($M)

 
6
80
5
60
4
Changes in payment
For each unit new claim of $1, the first 3 40
entries of CR, CQR, CQ2R, and CQ 3R are 2
0, 0.669, 0.127 and 0, and these enter 20
1
the ‘Paid’ state in subsequent time pe-
riods. Thus the first claim of $95.0K (in 0
the first month) gives rise to payments 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
of $0, $63.6K, $12.1K and $0 in subse- Claim number
quent months. Similarly, the second Figure 2: Claims data – case example B

484 o rga n i za t i o n , te ch n ol o g y a n d ma na ge m e n t i n co nst r u c t i o n · an international journal · 4(2)2012

 tal payment was less at approximately Claims According to Project and constant as project activities are be-
$903.1K. If this payment scenario had Cumulative Expenditure ing initiated and the project resources
been anticipated, the contractor could A piecewise linear approximation to a mobilised, a middle phase where expen-
have increased claim mark-up in order project cumulative expenditure or proj- diture is high and constant and contrac-
to improve its net cash flow. Such pay- ect S curve will cover most situations. tors could expect to submit claims of
ment information, if available from past Each straight-line portion represents a similar amounts regularly, and a final
projects, can be used to estimate pay- period of claims of constant but differ- phase where expenditure and claims are
ments on future projects. ing amounts. One, two or more straight- low and constant as the project winds
Below, a project S curve is approx- line segments may be appropriate, de- up (Blyth and Kaka, 2006).
imated by piecewise linear portions pending on the fluctuation of claims The data from three projects are
equivalent to claims constant in time over the project duration. The textbook shown here to demonstrate typical
but of differing magnitudes, in order project S curve might be approximated claims practices, and how the cumula-
that cash inflow estimation can be read- by three straight lines - an initial phase tive claims plots may be approximated
ily obtained. where expenditure and claims are low by multiple straight-line segments,
where the number of segments may
12 160 vary from project to project.
Case example B: the construction of
140 a 7 km two-lane grade separated road.
10
41 progress claims were made over a
120
total duration of 32 months (Figure 2).

Cummulative amount ($M)

8 Because of the peak claims either side
Claim amount ($M)

100
of small claims, the cumulative claim
6 80 schedule of this project may require ap-
proximating by several segments.
60
4 Case example C: the construction of
40 a rural highway including earthworks,
drainage, pavements, road furniture
2
20 and traffic management. The 22 prog-
ress claims are shown in Figure 3. This
0 cumulative plot might be approximated
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
by several segments, or more severely
Claim number by one segment.
Figure 3: Claims data – case example C Case example D: the refurbishment
of a city building with total cost of ap-
10 60 proximately $60M and duration of ap-
proximately 20 months. The cumulative
9
claims given in Figure 4 might be ap-
50
8 proximated by two straight segments
either side of the middle of the project.
Cummulative amount ($M)

7
40 The number of straight-line seg-
Claim amount ($M)

6
ments assumed to represent cumula-
5 30 tive expenditure is at the discretion of
the contractor. Stylised assumptions, in
4
20 order to simplify the calculations, how-
3 ever might be in terms of:
2 XX One straight-line (constant slope) seg-
10
ment over the entire project.
1
XX Two straight-line segments where the
0 change occurs near project midpoint.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
XX Three straight-line segments with the
Claim number larger claims in the middle part of
Figure 4: Claims data – case example D the project.

hanh tran · da vi d g . ca r m i cha e l · c o n t r a c t o r ’ s f i n a n c i a l e s t i m a t i o n b a s e d o n o w n e r  … · pp 481 - 489 485

 2.0

1.5 Payment

1.0

0.5
Amount ($)

0.0

-0.5

-1.0

-1.5 Claim

-2.0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37

Time (months)

Figure 5: Case example A extension; payment-claim relationship based on unit claims for the first 9 and last 9 months,
and twice this for the middle 18 months

Nonlinear segments can also be 3 and 4 are $0, $0.669, $0.127 and $0, out the project. The mark-up of 17.6% is
assumed in place of linear segments, respectively. The payment-claim rela- not high enough to give a positive pres-
for example by using quadratic or ex- tionship is plotted in Figure 5. ent value as the contractor would like. The
ponential functions. Numerical experi- minimum mark-up that needs to be ap-
ments conducted by the authors show Present Value plied in order to have a non-negative pres-
that the difference in the results be- Let the net cash flow, the difference be- ent value is of interest to the contractor.
tween linear and nonlinear assumptions tween the payments and the expendi- Figure 6 shows a range of mark-
is negligible. In the steady state there ture at each time period, i = 0, 1, 2, ... ups and the associated present value
is almost no difference in the values be xi. The present value (PV) is the sum amounts. It is seen that the contractor
of the matrix NR or the vector CNR be- of the discounted xi, needs to adopt at least a 27% mark-up
tween the assumptions of linear or non- (value at the intersection of the present
linear segments. Accordingly the extra value plot and the horizontal axis) in or-
accuracy that might be thought possible   der to have a non-negative present value.
through the use of nonlinear approxi-
mations is not there; as well, it comes Effect of delaying payments to creditors
with increased burdens of mathematical where r is the monthly discount rate on contractor’s cash flow
understanding and computational load. and m extends until the last payment In order to improve present value, the con-
is received. tractor may consider the option of delaying
Estimation for a Future Project Consider the values as in Figure 5. payments to creditors such as subcontrac-
– Case Example A Extension The steady state payment to the ‘Paid’ tors and suppliers. This delays cash out-
Assume that the owner payment practices state is $0.796. Since the actual cash flows. Different time lags in payments to
of case example A apply, but now add the outflow each month is 85% of each claim, creditors can be examined by shifting the
following new (future) project specifics. the contractor’s net cash flow in each of cash outflows to the right.
There are 36 monthly progress claims, the first 36 months is negative. The non- Consider the situation (in the same
where the first 9 claims, the next 18 claims discounted total payment is $42.98 while case example) in which the contractor de-
and the last 9 claims have ratios of 1:2:1. the non-discounted total expenditure is lays its cash outflow by one month, then
The claims submitted include a mark-up $45.90. Based on a monthly discount the present value for a 17.6% mark-up be-
(17.6%) to account for overheads; the ac- rate of 1%, the present value of the net comes -$2.50. To make the present value
tual spending of the contractor is 85% of cash flow is -$2.90. non-negative, a 26% mark-up is required.
what it is being claimed. The contractor Delaying cash outflow by a further month,
pays for the work as it is done irrespective Effect of mark-up on present a 25% mark-up is required to bring the
of getting any payment from the owner. value present value to a positive amount ($0.13).
For each unit or $1 claim, the change The above analysis shows that the contrac- This may assist the contractor’s bid to be
in payments in subsequent months 1, 2, tor has a negative net cash flow through- more competitive.

486 o rga n i za t i o n , te ch n ol o g y a n d ma na ge m e n t i n co nst r u c t i o n · an international journal · 4(2)2012

 Figure 7 shows how the present value Typical owner payment Average payment in
changes for different time lags in paying behaviour Owner Type 2 months (standard
creditors, assuming a 25% mark-up. A study of the classification of owner deviation)
The effect on present value of delay- payment behaviour by Tran and Carmi- 1 0.54 (0.24)
ing the cash outflow is more apparent chael (2013) established that there are
2 0.27 (0.1)
when the discount rate is higher. With six main types of owners when char-
a discount rate of 1.5% per month, the acterised in terms of their late and in- 3 0.70 (0.15)
minimum mark-up required to have a complete payment histories. Owners 4 0.34 (0.13)
non-negative present value for delays are classified according to three pa-
5 0.76 (0.11)
of 0, 1, and 2 months is approximately rameters representing uncertainties
26%, 25% and 24%, respectively. in payments, namely, the proportion 6 0.80 (0.11)

The above example shows the con- of total amount paid within a certain
Table 2 Average payment in 2-month
tractor the impact of mark-up choices and time frame, the time following the sub-
period following claim submission,
creditor payment policies on its finances. mission of the claim to the initial pay-
as a proportion of total claimed
This may assist, for example in being ment made, and the consistency in the
amount, of the six representative
more competitive at tendering time, or promptness in responding to each in-
owner types of Tran and Carmichael
in administering funds during a project. dividual claim.
(2013)

4.00 Accordingly, it is shown that owners

with incomplete payment histories fall
within one of six levels of practice: from
2.00 poor – Type 1 to excellent – Type 6 (Tran
and Carmichael, 2013). The anticipated
Present value

0.00 payment in terms of proportion of to-

tal claimed amount, for example in 2
10 20 30 40 months following claim submission,
-2.00 for each typical owner type is shown
in Table 2.
For owners Type 1, Type 2 and Type 4,
-4.00
given that the steady state paid amount
in the 2-month allowance is no more than
-6.00 Percent mark-up 60% of the claim, the contractor’s real
expenditure should be lower than 60%
Figure 6: Effect of mark-up on PV - case example A extension of its claimed amount in order to have
a positive monthly net cash flow. This
0.60
implies a mark-up of more than 100%.
Therefore, such owners are not desirable
0.40 to work for. Owners Type 5 and Type 6
have steady state payments equal to 76%
0.20 and 80% of the claimed amount, respec-
0 month 1 month
Present value

tively. Hence mark-ups of at least 31.5%

0.00
and 25%, respectively, are required in or-
2 months 3 months
-0.20 der to have a positive present value when
working with these owner types. Owner
-0.40 Type 3 may also be suitable to work for,
but a contractor might also simultane-
-0.60
ously consider other practices such as
-0.80 Time of payment to creditors (month) front-end loading or up-front payments.
The classification of owner payment
Figure 7: Present value based on different payment time lags to creditors; behaviour allows contractors to per-
25% mark-up, 1% monthly discount rate - case example A extension form an analysis based on the identi-

hanh tran · da vi d g . ca r m i cha e l · c o n t r a c t o r ’ s f i n a n c i a l e s t i m a t i o n b a s e d o n o w n e r  … · pp 481 - 489 487

 fied type of the owner, derived from the 4. For each claim, calculate the values ent project phases as demonstrated
contractor’s own experiences or others’ of CR, CQR, CQ2R, ..., CQn-1R, where C in case examples B, C and D. The con-
experiences. The requirement of hav- = [c, 0, ..., 0] is a 1×n vector and c is tractor may also examine different op-
ing specific historical owner payment the amount of the claim. These are tions in claim submission schedules,
data can be eased, yet the result of the payments in the weeks or months fol- taking into account any possible front-
analysis remains practical. For exam- lowing claim lodgement. end loading. The methodology remains
ple, consider a contractor working with 5. Add claim payments to the cash out- the same.
an owner Type 6. Based on the antici- flow diagram to produce a complete The analysis is not only applicable
pated payment practices of this owner cash flow diagram. to owners with incomplete payment his-
(as given in Table 2) and the assumed fu- 6. Perform a present value analysis. Ex- tories, but also applicable to complete
ture project scenario as in Figure 5, the amine changes in assumptions on payment situations as identified in Tran
contractor is advised to adopt at least mark-up, discount rate and payment and Carmichael (2013). Based on knowl-
a 25.5% mark-up, assuming payments time lag to creditors in order to as- edge of the timing of payments from
to creditors are not delayed. The mark- sist decision making on tendering the owner, the contractor can perform
up can be reduced to 23.9%, 22.7% policy and/or project administration the same analysis to estimate the in-
and 21.5% respectively for 1-, 2- and practices. crements in payments following claim
3-month payment time lags to creditors. lodgement and feed this into the cash
The contractor, based on this analysis, Conclusion flow diagram. Because claims are paid
can also modify its cash flow diagram The paper provides a practical way for a completely, the timing of payments may
by unbalancing its claims schedule to contractor to perform financial calcula- not largely affect the present value, but
give a further reduction in the mark-up, tions based on past payment experience it still gives very useful information
thereby improving its competitiveness. with an owner, or estimates of an own- about the monthly/weekly cash flow of
er’s payment practices. The contractor the contractor.
Summary of the Approach for a is able to forecast future cash flows and Future research. The analysis pre-
Contractor hence potential project profitability. The sented in this paper could be extended
The above development is summarised method is best applied pre-tender when by considering different claim sched-
for the purpose of being implemented by simple and quick cash flow estimates ules made by the contractor in an up-
contractors (as well as subcontractors, are required. It can also be used dur- coming project. Instead of assuming
suppliers and consultants when deal- ing a project to understand the cash the cash outflow being constant over
ing with others higher in the contractual position of the contractor, or to adjust certain project phases, cash outflow
chain). The method requires no more than claims practices. estimated from actual project sched-
a summary or estimate of outstanding For each owner type, the method ules could be used to estimate payment
project money against time after claim allows contractors to: portions and timing. Another extension
lodgement from a past project. All cash XX Calculate the payment expectancy of the research could be incorporating
flow and present value calculations can for individual claims, including incre- probabilistic cash flow forecasts into
be readily done using a spreadsheet. ments in payments in weeks/months the analysis, taking into account un-
1. Decide on a relevant time period and following claim lodgement. certainty associated with discount rate,
how many time periods must pass be- XX Generate a cash flow diagram by look- project schedules and investment life
fore a claim is conceded as needing ing at a series of claims. spans. The whole analysis could be inte-
resolution. Based on past projects or XX Perform a present value analysis of grated into a spreadsheet tool requiring
estimates, summarise the outstand- payments and expenditure, consid- only user inputs of a summary of owner
ing amounts against time since claim ering various possible time lags in historical payment data, and a future
lodgement. Estimate the entries of the cash outflow to creditors, discount cash outflow schedule. The spread-
matrix P in the Markov chain formula- rates and mark-ups, and hence de- sheet tool would allow the contractor
tion (Carmichael and Balatbat, 2010). cide on the most suitable bidding and to examine the effect of tardy payments
2. Calculate the submatrices Q and R. claim practices. from the owner, possible changes in
3. Generate a cash outflow diagram The method permits a number of dis- discount rates and payment policies to
based on the project schedule. Sim- cretionary parameters including choice creditors on its net cash flow and net
plify the cash outflow diagram by al- in time periods, and choice in time lags present value. The classification of typi-
lowing constant claims over project in payment to creditors. The cash out- cal owner payment behaviour could be
phases, if detailed estimation is not flow calculations may be simplified by incorporated in the analysis tool for
available. allowing constant claims over differ- quick and simple estimation.

488 o rga n i za t i o n , te ch n ol o g y a n d ma na ge m e n t i n co nst r u c t i o n · an international journal · 4(2)2012

References Frydman, H., Kallberg, J. G. and Kao, D. (1985), Tran, H. and Carmichael, D. G. (2012a), The
Barkman, A. I. (1977), Estimation in accounting Testing the adequacy of Markov chain and Likelihood of Subcontractor Payment;
and auditing using Markov chains, Mover-Stayer models as representations of Downstream Progression via the Owner and
The Journal of Accountancy, 144 (6), 75-79. credit behaviour, Operations Research, Contractor, Journal of Financial Management
Blyth, K. and Kaka, A. (2006), A novel multiple 33 (6), 1203-1214. of Property and Construction, 17 (2) ), 135-152.
linear regression model for forecasting Hwee, N. G. and Tiong, R. L. K. (2002), Model of Tran, H. and Carmichael, D. G. (2012b),
S-curves, Engineering, Construction and cash flow forecasting and risk analysis for Stationarity of the Transition Probabilities
Architectural Management, 12 (1), 82-95. contracting firms, International Journal of in the Markov Chain Formulation of Owner
Brand, M. C. and Uher, T. (2010), Follow-up Project Management, 20 (5), 351-363. Payment Histories on Projects, Australian
empirical study of the performance of the Ho, S. P. and Liu, L. Y. (2003), How to Evaluate and New Zealand Industrial and Applied
New South Wales construction industry security and Invest in Emerging A/E/C Technologies Mathematics Journal, 53 (C69-C89).
of payment legislation, International Journal of under Uncertainty, Journal of Construction Tran, H. and Carmichael, D. G. (2013), A
Law in the Built Environment, 2 (1), 7-25. Engineering and Management, 129 (1), 16-24. Contractor’s Classification of Owner Payment
Carmichael, D. G. (2000), Contracts and Kaka, A. P. (1996), Towards more flexible and Practices, Engineering, Construction and
International Project Management, accurate cash flow forecasting, Construction Architectural Management 20 (1)., accepted for
A A Balkema, Rotterdam. Management and Economics, 14 (1), 35-44. publication, to appear 2013.
Carmichael, D. G. (2002). Disputes and Kaka, A. P. and Price, A. D. F. (1991), Net cash Tran, H., Carmichael, D. G. and Balatbat,
International Projects, A A Balkema, flow models: are they reliable, Construction M. C. A. (2011). Tree form classification of
Rotterdam. Management and Economics, 9 (3), 291-308. owner payment behaviour. Proceedings of the
Carmichael, D. G. and Balatbat, M. C. A. (2010), Kaka, A. P. and Price, A. D. F. (1993), Modelling 4th International Conference on Construction
A contractor’s analysis of the likelihood standard cost commitment curves for Engineering and Project Management, Sydney
of payment of claims, Journal of Financial contractors’ cash flow forecasting, 16-18 February 2011, University of New South
Management of Property and Construction, Construction Management and Economics, Wales, Sydney, Australia.
15 (2), 102-117. 11 (4), 271-283. van Kuelen, J. A. M., Spronk, J. and Corcoran,
Chen, H. L. and Chen, W. T. (2000). An interactive Kallberg, J. G. and Saunders, A. (1983), Markov A. W. (1981), On the Cyert-Davidson-Thompson
cost-schedule/payment schedule prototype chain approaches to the analysis of payment doubtful accounts model, Management
integration model for cost-flow forecasting behaviour of retail credit customers, Financial Science, 27 (1), 108-112.
and controlling. 17th International Association Management, 12 (2), 5-14. Wort, D. H. and Zumwalt, J. K. (1985), The trade
for Automation and Robotics in Construction. Kenley, R. and Wilson, O. D. (1986), A construction discount decision: a Markov chain approach,
IAARC, Taipei, Taiwan. project cash flow model—an idiographic Decision Sciences, 16 (1), 43-56.
Chen, H. L., O’Brien, W. J. and Herbsman, approach, Construction Management and Wu, J., Kumaraswamy, M. M. and Soo, G. K.
Z. J. (2005), Assessing the accuracy of Economics, 4 (3), 212-232. L. (2011), Dubious benefits from future
cash flow models: the significance of Kenley, R. and Wilson, O. D. (1989), A construction exchange: an explanation of payment arrears
payment conditions, Journal of Construction project net cash flow model, Construction from ‘continuing clients’ in Mainland China,
Engineering and Management, 131 (6), Management and Economics, 7 (1), 7-18. Construction Management and Economics,
669-676. Mavrotas, G., Caloghirou, Y. and Koune, J. (2005), 29 (1), 15-23.
Corcoran, A. W. (1978), The use of exponentially- A model on cash flow forecasting and early
smoothed transition matrices to improve warning for multi-project programmes:
forecasting of cash flows from accounts application to the Operational Programme
receivables, Management Science, for the Information Society in Greece,
24 (7), 732-739. International Journal of Project Management,
Cyert, R. M., Davidson, H. J. and Thompson, 23 (2), 121-133.
G. L. (1962), Estimation of the Allowance Navon, R. (1995), Resource-Based Model for
for Doubtful Accounts by Markov Chains, Automatic Cash-Flow Forecasting, Construction
Management Science, 8 (3), 287-303. Management and Economics, 13 (6), 501-510.
del Cano, A. and de la Cruz, M. P. (2002), Park, H. K., Han, S. H. and Russell, J. S. (2005),
Integrated Methodology for Project Risk Cash Flow Forecasting Model for General
Management, Journal of construction Contractors Using Moving Weights of Cost
engineering and management, 128 (6), Categories, Journal of Management in
473-485. Engineering, 21 (4), 164-172.
El-Adaway, I. H. and Kandil, A. A. (2009), Russell, J. S. and Zhai, H. (1996), Predicting
Contractors’ Claims Insurance: A Risk contractor failure using stochastic dynamics
Retention Approach, Journal of construction of economic and financial variables, Journal of
engineering and management, construction engineering and management,
135 (9), 819-825. 122 (2), 183-191.

hanh tran · da vi d g . ca r m i cha e l · c o n t r a c t o r ’ s f i n a n c i a l e s t i m a t i o n b a s e d o n o w n e r  … · pp 481 - 489 489