An Evaluation of Cost Performance of Public Projects: Case of

Botswana

D K CHIMWASO
Department of Architecture and Building Services
Private Bag 0025, Gaborone
Botswana
GFKLPZDVR#KRWPDLOFRP

Abstract

Improving cost performance of public projects, still remains one of the challenges facing the construction
industry in developing countries. Research has found that there are many cases of cost overruns as
compared to projects that have been completed within budget. This paper seeks to evaluate the cost
performance of public projects in Botswana, and tries to identify factors that influence construction cost
overruns.

The study was based on a questionnaire survey among professionals of the construction industry. The
results, together with empirical data from ten completed projects have been presented. Five significant
factors, that influence construction cost overruns, have been identified, and they include incomplete design
at time of tender, technical omissions at design stage and contractual claims. These factors have further
been classified under such categories as variations and contractual claims, according to the format of final
account reports.

The paper recommends that there is need to identify significant factors that may influence construction cost
overruns and deal with them from the inception of the project. This will result in significant decrease in the
occurrence of cost overruns and improve cost performance of projects. Although the research concentrates
on public projects in Botswana, the findings should be valuable to other developing countries.

Keywords: Contract sum, Cost performance, cost overruns, client/customer satisfaction, factors.

INTRODUCTION

One of the challenges facing the construction industry in developing countries is the chronic problem of
construction cost overruns. Under normal circumstances it would be possible for a complete set of
drawings and specifications to be made available to the Quantity Surveyor, who would prepare fully
described and accurate bills of quantities for tendering purposes, The Aqua Group (1992). This would
lessen changes during the construction stage, as the contractor would have all the necessary information on
which to base the price. However, such a situation is not always the case, (Knowles, 1986; Fisk, 1988).

As a result the contract sum cannot always be regarded as a firm price. Even where the work is ordered on
“Fixed price contract” basis, there will still be remeasurements of provisional works, adjustment of prime
cost and provisional sums, and variation orders to contend with. The cost arising from these unforeseen
circumstances is the sure cause for the adjustment of the contract sum, (Ashworth, 1988; Seeley, 1997;
Kwakye, 1997)
It happens, therefore, that the initial tender price is not what is finally paid. This is because there are many
factors that influence cost overruns. This, without saying, results in poor cost performance of construction
projects, leaving many a customer unsatisfied.

It should be noted that a favourable cost performance of any given project depends on the effectiveness of
the contract sum.

The word “effective” according to the Oxford Advanced Learner’s Dictionary, means having an effect or
producing an intended result. It also means ‘actual’, like in actual cost of construction works.

In construction, however, it is difficult to talk about the actual construction cost, that is, the money the
client pays the contractor for carrying out the work, Kwakye (1997), until the project has been completed
and the final account agreed upon. For this reason, the anticipated final construction cost is always referred
to as the projected or estimated final account.

Despite this state of uncertainty, the client will always want to know how much the project would cost,
(Ashworth, 1988; The Aqua Group, 1992; Seeley, 1997). As Ashworth argues out, even where the project
is to be carried out as a matter of urgency the cost may be of less importance but cannot be easily ignored,
(ibid). It should be noted that construction cost, like the quality of work, is one of the Key Performance
Indicators (KPIs), or performance measurement standards, (Kerzner, 1998; Martin, 1999). This poses a
challenge to the professionals of the construction industry. A challenge to improve productivity and cost
performance of projects to the satisfaction of the customer.

This is particularly becoming imperative, more so that cost and time overruns have given the construction
industry a poor image, Radujkovic (1999). Radujkovic cites a recent study in which only 16% of the
projects were considered successful, that is, completed on time, within budget, and to specification. He also
cites another report that recorded substantial cost overruns in 63% of the construction projects that were
studied.

Although local literature is hard to come by, experience has shown that the construction industry in
Botswana, like in other developing countries, is not free from incidents of cost overruns. The question is
what should be done so that cost performance of public projects can be improved and how?

Usually, cost overruns are loosely blamed on lack of cost control. However, this is said to stem from a poor
definition of cost control that focuses on only monitoring of costs and recording of data. Whereas, an
effective cost control system is dependent on the original plan of works against which performance will be
measured, Kerzner (1998).

This paper seeks to evaluate the cost performance of public projects in Botswana and sets out to identify
and rank the factors that influence cost overruns. Research findings are presented together with empirical
data from completed public projects. In discussing cost overruns, this paper limits itself to the construction
cost as represented by the tender amount, which once accepted becomes the contract sum. Previous
research has attempted to discover the reasons for the disparity between the tender sum and the final
account, although without much success, Ashworth (1988). This study is of great importance in that it
raises the level of awareness and gives the design team a sense of preparedness from the inception of the
project.

The paper has been organised as follows: first, there is a literature review on the problem of cost overruns,
which will be followed by identification of the factors that influence cost overruns. Second, the objectives
of the study, the method of data collection, and the structure of the questionnaire employed are described.
This will be followed by presentation of the research findings together with empirical data from ten
completed public projects in Botswana. Third, the results of the questionnaire survey together with the
empirical data from completed projects will be examined and discussed. The factors that influence cost
overruns will also be ranked in order of significance. The significant factors will then be classified in
categories according to the format of final account reports, and further ranked in order of importance.
Finally, conclusions will be made together with recommendations.

THE PROBLEM OF COST OVERRUNS

There have been many research projects on cost and time overruns, (Kaming, et al., 1997; Cox, et al., 1999;
Radujkovic, 1999).

These research findings have one thing in common. It was found that there were more cases of cost
overruns than time overruns. This makes the problem of cost overruns to be of great significance.

Most of the previous studies had concentrated on a few factors that influence time and cost overruns, such
as design changes which has been reported to account for 5 – 8% of the total project cost, Cox, et al (1999).
This selective approach leaves out many factors that one encounters in the final account reports. The best
approach should, probably, be to investigate most of them and rank them accordingly. Although some of
the factors may seem to be insignificant on one project, they may prove to be significant on another project,
as conditions are not always the same.

It would appear, there will always be need for debate and further research because of the chronic problem of
construction cost overruns. As one arbitrator once observed, it is difficult to imagine a building contract,
which proceeded to completion without delay or variation whatsoever, Knowles (1986). However, this
does not mean that there are no building projects that have been completed within budget. The concern is
that such ideal situations are rare, (ibid.; Fisk (1988). Research has found that there are more building
projects that had cost overruns than those which were completed within budget, Radujkovic (1999). The
scenario of cost overruns has been blamed on the many factors that influence construction cost overruns,
Kaming, et al (1997).

IDENTIFYING FACTORS THAT INFLUENCE COST OVERRUNS

Previous research has attempted to discover reasons for the disparity between the tender sum and the final
account, Ashworth (1988). This section identifies the factors that influence cost overruns. As has already
been stated, this study considers cost overruns in relation to the tender amount, which once accepted
becomes the contract sum. As such only factors that influence such cost overruns were identified.

Four factors were identified from the existing research findings, (Kaming, et al., 1997), Cox, et al., (1999),
and Radujkovic, 1999). These are as follows:

1. Design changes;
2. Inadequate planning;
3. Unpredictable weather conditions; and
4. Fluctuations in the cost of building materials.

To broaden the investigation, it was decided to complement the above list of factors with other factors
gleaned from the final account reports. These were compared with the factors from the existing research
findings, and a final list of 18 factors was prepared. These were then divided into two groups of seven
critical factors and nine other factors, which are usually ignored, but perceived to be of equal significance.
The critical factors are the ones that were considered to be significant, and these incorporated the ones that
were identified to be significant from the previous research findings. The said two lists of factors which
were identified are as follows:

List of critical factors:

1. Incomplete design at the time of tender.
2. Additional work at client’s request.
3. Changes in client’s brief.
4. Lack of cost planning/monitoring during pre-and- post contract stages.
5. Site/poor soil conditions.
6. Adjustment of prime cost and provisional sums.
7. Remeasurement of provisional works.
8. Logistics due to site location.
9. Lack of cost reports during construction stage.

List of other factors, which are usually ignored:

1. Delays in issuing information to the contractor during construction stage.
2. Technical omissions at design stage.
3. Contractual claims, such as, extension of time with cost claims.
4. Improvements to standard drawings during construction stage.
5. Indecision by the supervising team in dealing with the contractor’s queries resulting in delays.
6. Delays in costing variations and additional works.
7. Omissions and errors in the bills of quantities.
8. Ignoring items with abnormal rates during tender evaluation, especially items with provisional
quantities.
9. Some tendering maneuvers by contractors, such as front-loading of rates.

The above factors were then used in the formulation of the questionnaire.

OBJECTIVES AND METHODOLOGY

The objective of the study was to identify the factors influencing cost overruns, rank them in order of
significance to raise the level of awareness. The objective was to be achieved through a questionnaire
survey and an evaluation of final account reports for ten completed public projects to provide empirical
data.

From the existing research findings and final account reports, it was possible to identify the factors that
influence cost overruns. These were organised into a questionnaire. The questionnaire was designed to
enable respondents to add any other factor that they considered necessary for inclusion to the list of 18
factors. The questionnaire survey was supported by, informal interviews and discussions with some of the
respondents.

DATA COLLECTION

Responses to a questionnaire were received from forty-six professionals of the construction industry,
comprising architects, engineers, quantity surveyors, and project managers. The respondents comprised
consultants in the private sector and the professionals in the public service with over five years’ experience
of the Botswana situation.

Fluctuations in the cost of labour and materials were not listed among the factors. It was, however,
expected that respondents would bring it up, and this is reported under any other factors in table 1.

RESULTS, DATA PRESENTATION AND DISCUSSION

The first question was to establish whether construction cost overruns is a problem in Botswana. All
respondents considered it a problem that needs to be addressed. However, they did not consider all the
above listed factors to be significant in as far as their experience is concerned. Also they did not consider
tendering maneuvers by contractors, such as front-loading of rates to gain as much money as possible
during the early stages of construction, to be a problem in Botswana. Nevertheless, all the factors that were
listed were considered to be contributory factors that should not be ignored on any given project.

Four factors were considered to be critical (see table 1). These were the ones considered by at least 50% of
the respondents.

Table 1. Critical factors

Factor Respondents
Incomplete design at time of tender 78%
Additional work at client’s request 57%
Changes in client’s brief 52%
Lack of cost planing/monitoring 50%
Site/poor soil conditions 41%
Adjustment of p. c. and provisional quantities 26%
Re-measurement of provisional quantities 24%
Logistics due to site location 20%
Lack of cost reports 13%
Any other 2%

Three factors featured prominently in the list of other factors that are normally ignored. Again, these were
the ones that were considered by at least 50% of the respondents.

Table 2. Other factors

Factor Respondents
Delays in issuing information (Construction stage) 72%
Technical omissions at design stage 57%
Contractual claims (such as, extension of time) 50%
Improvements to standard drawings 46%
indecision by the supervising team 41%
Delays in costing variations 33%
Omissions & errors in bills of quantities 24%
Front-loading of rates 22%
Items with abnormal rates 20%

Table 1 and 2 above, shows in general terms the areas of concern. Respondents were then asked to state the
increase in cost over and above the contract amount on most of their projects at the final account stage.
This is recorded in table 3.

Table 3. Degree of cost overruns on most projects

Percentage increase Respondents
Less than 5% 26%
5% to 15% 48%
16% to 30% 9%
31% to 45% 2%
46% to 60% 4%
61% to 75% nil
76% to 100% 2%
over 100% nil

The main causes for the above reported cost overruns were given as:

1. Incomplete design at time of tender.

2. Additional work at the client’s request.
3. Technical omissions at design stage.
4. Adjustments of prime cost and provisional sums.
5. Contractual claim, that is, extension of time with cost claims.

It is clear from the above factors that the first three would fall under variations. This gives three prominent
categories, in relation to the format of final account reports as outlined below. However, the list does not
tie up with the one derived at after examining the empirical data which lists four categories, including
remeasurement of provisional works and fluctuations in the cost of labour and materials as will be seen
later.

1. Variations.
2. Adjustment of prime cost and provisional sums.
3. Contractual claims, that is, extension of time with cost claims.

Empirical data

Ten randomly selected projects were examined to validate the above research findings. The investigation
involved examining the respective final accounts to assess the influences of the various factors on the
contract sum. It is clear that the empirical data in table 4 below, compares well with the results of the
questionnaire survey as recorded in table 3, regarding the degree of cost overruns on most projects. Cases
of cost overruns applied on both small and large projects. As a result it could not be determined whether
there was sufficient planning of design works for big projects as compared to the small ones.

Causes for the above stated cost overruns, over and above the contract sum, were remeasurement of
provisional works (3.28%), variations (6.34%), fluctuations in the cost of labour and materials (3.34%), and
contractual claims, that is, claims for extension of time with cost (2.70%). It is evident from the above
categories that the list of cost overrun influences do not tie up with the one arrived at from the
questionnaire. The differences could be due to various circumstances as will be explained later in the
analysis of the above categories. However, judging from the empirical data, four categories do emerge,
which excludes the adjustment of prime cost and provisional sums. These categories are as follows:

1. Variations.
2. Remeasurement of provisional works.
3. Fluctuations in the cost of labour and materials.
4. Contractual claims, that is, claims for extension of time with cost.

By classifying the various influencing factors under the above categories, narrows the problem and helps to
deal with it effectively. It also helps to consider other factors under the respective categories before
dismissing them as insignificant on any given project. It is not realistic however to believe that all the
factors that influence cost overruns can be brought under control. Nevertheless, it would be sensible to
identify them and try to reduce their influence, Al-Khalil and Al-Ghafly (1999). As Radujkovic (1999)
points out, by identifying and analysing the factors that influence cost overruns, it will raise the quality
level of preparation and project control, and this will result in significant decrease in the occurrence of cost
overruns.
 Table 4. Case study - sample projects
Project Completion date Cost overrun
Offices block, Kazungula 1989 7%
Office block, Maun 1990 10%
Office block, Makopong 1991 -13%
Offices block, Selebi-Phikwe 1991 21%
Office block, Goodhope 1994 -24%
Office block, Maun 1994 2%
Office block, Gaborone 1995 14%
Office block, Kasane 1996 15%
Offices block, Gaborone 1997 7%
Aircraft hanger, Gaborone 1998 -10%

The above table shows that seven out of ten projects had cost overruns. Although this may not be the most
reliable conclusion, given the small sample, it however indicative of the problem of cost overruns on public
projects in Botswana.

ANALYSIS

Variations

Variations are change orders. These design changes may be ordered at the client’s request, or the architect
or engineers decision to rectify any errors or omissions in the quality or volume of work shown on contract
drawings, and as described or referred to in the bill of quantities, The Aqua Group (1986). Variations have
a double effect. First, they may contribute to cost overruns and second, they may also affect the progress of
the works, resulting in claims for extension of time with cost by the contractor concerned, in the case of
increased scope of work.

From the questionnaire and empirical data analysis, variations could be ranked as having the most
significant influence on cost overruns. However, it would be difficult to rank the remaining three
influencing factors in order of significance as their influences are closely related. All the four factors
should, however, be regarded as significant factors when considering the factors that may influence cost
overruns on any given project. The significance of these factors is supported by findings of the previous
research projects reported earlier in this paper, that is, design changes, fluctuations, and insufficient
planning of design work.

As can be seen, both design changes and insufficient planning of design work can be classified under
variations. There are many factors that can be classified under variations including additional at the client’s
request. This makes variations one of the most significant categories to be considered in the
implementation of any project. How, if one may ask, can the change orders by the client be avoided? The
answer is by understanding the client’s requirements from the briefing stage and to come up with a concise
brief. This was the feeling of most of the professionals who were interviewed.

From the sample of ten projects reported in this paper, it was difficult to identify any variations ordered at
the client’s request for most of them. However, there seem to be a change in the presentation of variation
orders as one examines some of the latest final accounts. There is a clear distinction between instructions
issued by the architect and any changes ordered at the client’s request. This is because the client’s
professional advisors have, in recent years, been asked to justify why they had issued variation orders. This
approach could improve cost performance of projects, as the professionals concerned will be more careful
during the design stage.
REMEASUREMENT OF PROVISIONAL WORKS

The substructure, plumbing, and external works are usually measured as provisional in the bill of quantities,
subject to remeasurement at a later stage during the construction of the works. Remeasurement becomes
necessary due to, but not limited to, the following reasons.

1. Unforeseen circumstances such as poor soil conditions which may necessitate re-designing of the
foundations. For example, there has been a case in Botswana where contractors were asked to dig up
pits of up to one meter deep and fill them up with imported material before constructing the
foundations.
2. Lack of details at tender documentation stage.
3. Where the foundations and roadwork have not been designed in accordance with the soil investigation
report, or where such a report had given misleading information or lack of it.
4. Substantial rock in foundations than envisaged which may require expensive blasting methods.

Remeasurements are inevitable on any project. However, if a project is well planned and detailed, chances
are, there will be fewer items to be remeasured. From the empirical data analysis of the projects examined,
remeasurements contributed an average cost overruns of about 3% of the contract sum. Considering
individual projects, it can be reported that some of them recorded cost overruns of up to 6% of the contract
sum as a result of remeasurements, while others recorded savings. This shows that it is possible to have
savings if the design is well developed and detailed, [Fisk (1988), The Aqua Group (1992)].

ADJUSTMENT OF PRIME COST AND PROVISIONAL SUMS

In Botswana, especially with public works, estimated values of prime cost sums are included in the bill of
quantities to cover electrical and mechanical installations by nominated sub-contractors. The work is then
tendered for separately, at a later stage. From the empirical data analysis one could see cases of either
underestimating or overestimating. Both practices are not good and are misleading. Provisional sums may
equally present problems. Provisional sums are included in the bills of quantities to cover work to be
executed by statutory authorities such as power and water connections. Provisional sums are also included
for work of which the full extent cannot be determined at the time of preparing the bills of quantities,
Seeley (1997).

From the empirical data analysis, it was noted that some of the projects that reflected savings had a lot of
provisional sums, which were later omitted from the contract. It was also evident that there were increases
in the value of measured works despite the savings arising from the provisional sums. This shows that there
could be possibilities of cost overruns even where savings are reported. A carefully planned project should
have less provisional sums.

FLUCTUATIONS IN THE COST OF LABOUR AND MATERIALS

Where a fluctuation clause is inserted in the contract documents, it allows the contractor to claim for the net
increase in the cost of labour and materials after the date of tender. Most respondents did not consider this
to be of great concern, although it has been found to be a significant factor after evaluating the sample of
projects. This is probably because of the practice of ordering projects on “Fixed price contract” basis in
Botswana.

This practice shifts the responsibility to the contractor who should diligently allow in his tender the cost for
fluctuations in labour and materials. How effective this is could be a subject for another research.
Contractual claims

These arise from loss and expense claims, and legal costs due to contractual disputes, Kwakye (1997). Non
of the ten projects examined were taken to arbitration. Most of the claims for the projects studied, were for
loss and expense claim relating to extension of time with cost.

From the questionnaire survey and the public projects evaluated, contractual cost overruns resulting from
claims of extension of time with cost, can be attributed to variations that result in increased scope of works.

General comments

It can be seen that most of the above factors, chiefly, emanate from actions and omissions of both the client
and his professional advisors. In this regard one could argue that the first step towards minimising cost
overruns is to deal with the human factor, Radujkovic (1999). Cost overruns of 5 – 15% as presented in
table 4 above are considered to be substantial and can offset the client’s overall budget. As has already
been stated, there are many factors that influence cost overruns. This makes it difficult to control cost
overruns during the construction stage alone. Instead, there should be sufficient planning of the project at
the inception stage. Drawings and other tender documents should be well detailed before going out to
tender.

The construction industry is said to be essential to any country’s economic development, Construction
Review (2000). And governments are increasingly talking about sustainable economic activities.
Consequently, the construction industry, as an economic activity, needs to be sustainable. However, to
ensure a sustainable construction industry, there is need to harness all the resources, that is, financial,
material, and human resources. The approach should aim at safeguarding the client in terms of value for
money, resulting from good cost performance of projects. Medley (1996) specifies three stages of cost
management. This is during the design stage, the tender stage (by getting the best price), and during the
construction stage. This is necessary because cost overrun can occur at any stage of the construction
process, Kerzner (1998). This implies good cost management from inception to completion. This being the
case all the members of the design team must be involved from the inception stage, Ashworth (1988).

It is true that the efficient system of cost control in the construction industry is the responsibility of the
quantity surveyor, Ashworth [(1988), Seeley 1997)]. However, the alleviation of the problem of project
cost overruns should be regarded as a matter of teamwork. This is because the actions and omissions of all
the members of the design team are likely to affect the construction cost of the project, Ashworth (1983).
We learn from literature that there is growing pressure to improve performance of the construction industry
even in developed countries, Martin (1999). To this effect, seven key performance indicators (KPIs) have
been developed in the UK and are becoming the standard benchmarking measurements. The key
performance indicators include the construction cost. This confirms the importance of the construction cost
as an integral part of any project specification

CONCLUSION

An evaluation of ten public projects was carried out to assess their cost performance. The results have
shown that seven out of ten projects had reported cost overruns. The factors that influence cost overruns
have been identified and ranked in order of significance. These factors have further been classified under
categories according to the format of final account reports. By classifying them into categories, helps to
deal with them effectively. The four categories arrived at are: variations, remeasurement of provisional
works, contractual claims and fluctuations in the cost of labour and materials, with variations being the
most significant.

From the results of the questionnaire survey and the empirical data from the sample of ten projects, it can be
concluded that public projects in Botswana, like in other developing countries, are not free from cases of
cost overruns. This poses a challenge to the professionals of the construction industry. A challenge to
improve cost performance of projects to the satisfaction of the customer.

RECOMMENDATION

From the above findings one can make the following recommendations:

1. There is need to identify the factors that may influence cost overruns and deal with them from the
inception of each project. This will decrease the occurrence of cost overruns, Radujkovic (1999).
2. Insufficient planning of design works has been alluded to in this paper. Since design changes are a
result of insufficient planning, a careful study should be carried out to determine the appropriate time
scale in which to produce designs and other tender documents. This will help improve the quality of
tender documents and lessen changes during the construction stage.

ACKNOWLEDGEMENTS

The author would like to thank the respondents of the questionnaire survey, the anonymous reviewer for the
invaluable comments, and the Department of Architecture and Building Services for the sample projects.

REFERENCES

1. Ashworth, A (1983) Building Economics and Cost Control, Worked Solutions, Butterworths, London,
pp. 1 – 124
2. Ashworth, Allan (1988) Cost Studies of Buildings, Longman Scientific and Technical, Essex, pp. 1 –
122.
3. A-Khalil, Mohammed I and Al-Ghafly, Mohammed A (1999), Important causes of delay in public
utility projects in Saudi Arabia, Construction Management and Economics, Vol. 17, pp. 647 – 655.
4. Cox, I D, Morris, J P, Rogerson, J H and Jared, G E (1999), A quantitative study of post contract award
design in construction, Construction Management and Economics, Vol. 17, pp. 427 – 439.
5. Fisk, Edward R (1988) Construction Project Administration, Third Edition, John Wiley & Sons, New
York, pp. 433 – 452.
6. Harris, Frank and McCaffer, Ronald (1995) Modern Construction Management, Fourth Edition,
Blackwell Kerzner, Science, London, pp. 106 – 107
7. Harold (1998) Project Management A System Approach to Planning, Scheduling, and Controlling,
Sixth Edition, Jon Wiley & Sons, New York, pp. 785 – 825
8. Kaming, Peter F, Olomolaiye, Paul O, Holt, Gary D and Harris, Frank C (1997), Factors influencing
construction time and cost overruns on high rise project in Indonesia, Construction Management and
Economics, Vol. 15, pp. 83 - 94
9. Kwakye, A A (1997) Construction Project Administration in Practice, The Chartered Institute of
Building, England, pp. 11 – 12
10. Martin, Joe (1999) KPIs: key to opportunities for chartered surveyors, Chartered Surveyor Monthly
(CSM), September, pp. 28 – 29
11. Medley, Larry G Sr. (1996) The Lie-Cycle Perspective Managing Cost Before it Occurs, Cost
Engineering, Vol. 38, No. 10, pp. 35 – 36
12. Radujkovic, M (1999) The causes of Time and Cost Overruns in Construction Projects, Proceedings of
the CIB W55 & W65 Joint Triennial Symposium Customer Satisfaction: A focus for research and
practice,
13. Seeley, Ivor H (1997) Quantity Surveying Practice, Second Edition, Macmillan, London, pp. 240 – 241
14. The Aqua Group (1986) Construction Administration for Architects and Quantity Surveyors, Sixth
Edition, Collins Professional Books, London
15. The Aqua Group, (1992) Pre-Contract Practice for the Building Team, Eighth Edition, Blackwell
Science, London, pp. 9 - 107
16. Special Report, (2000) In: Construction Review, Vol. 11, No. 4, May, pp.10 – 2.