Modeling the Evolution of National Road Procurement Strategies
Monica ALTAMIRANO
PhD Student
Delft University of Technology (TUDelft)
Delft , 2628 BX, The Netherlands
M.A.Altamirano@ tudelft.nl
Abstract— Many transportation agencies have discovered that traditional highway contract
administration procedures and project delivery methods do not meet current demands. In
response, they are turning to alternative contracting. Four trends are perceived in road
management. First, with respect to project delivery, more and more projects are contracted for
the whole life cycle of the road. Second, contractors are given increasingly more freedom or
design space, as the indicators used for monitoring their work become less operational and more
performance based. Third, governments follow a dual track strategy; managing a portfolio of
directly and indirectly financed projects; dependent on the project characteristics. Fourth,
contracts are granted for longer term. These innovative forms of contracting are expected to
yield more flexibility in the road sector; more innovation, higher performance and consequently
lower costs while keeping up service levels on public values.
This paper presents how by using a combination of institutional economics theory and
engineering design theory, our aim is to build a systems dynamics model that can capture the
institutional context and is able to indicate what contracting practices are likely to occur and
which ones are likely to succeed in view of the meeting the public values and demands.
Keywords— System Dynamics, System Analysis, Road contracting, Procurement Strategy of
Roads.
1. Introduction
Many transportation agencies have discovered that traditional highway contract
administration procedures and project delivery methods do not meet current demands. Some of
the problems being faced by Road A dministrations around the world are; insufficient funds to
meet satisfactory levels of all roads (Miller 2000; Cox, Molenaar et al. 2002; Pakkala 2002),
little innovation (Manley and McFallan 2003), little value-added services for the client and a
general lack of integration between the phases of the road life cycle.
In response, they are turning to alternative contracting procedures to keep up with the
reconstruction and growth needed. Four trends are perceived in road management (A ltamirano
and Haraldsson 2005; Altamirano and Herder 2006). First, with respect to project delivery, more
and more projects are contracted for the whole life cycle of the road. Second, contractors are
given increasingly more freedom or design space (Herk, Herder et al. 2004), as the indicators
used for monitoring their work become less operational and more performance based (Cervera
and Minchin 2003). The third trend, concerns project financing in which private investors are
playing and increasingly higher role and governments follow a dual track strategy (Miller 2000);
managing a portfolio of directly and indirectly financed projects; dependent on the project
characteristics. Fourth, contracts are granted for longer term.
These innovative forms of contracting are expected to yield more flexibility in the road
sector; more innovation, higher performance and consequently lower costs while keeping up
service levels on public values such as mobility, safety and the environment. However, the
limited application of these innovative contracts and the continuous reform in all the different
dimensions of the procurement strategy (i.e. project financing, project delivery methods) do not
allow to reach consensus about the effects of these practices in terms of performance and
economic efficiency. Moreover, as the government agencies advance in the use of these
innovative procurement practices and its role become more that of a regulator and quality
monitor, than a direct operator of the road network; the more urgent is to research the possible
complementarities and tensions between the (technical) aspects of flexibility and those regarding
public values.
Consequently, the problem being faced by national road infrastructure systems and by many
other network bound infrastructures in the last decades can be pictured as an important trade-off
of “How to achieve the results expected from innovative contracting - mainly improvements in
efficiency and innovation-, while keeping the room for strategic behaviour (of contractors) and
the effects of it at the minimum possible?”
The paper presents the research framework being proposed for the analysis and study of
how different countries have given solutions to this problem, by applying and evolutionary and
systems dynamics approach. System Dynamics presents a great opportunity to understand the
evolution of these different trends because this technique allows the researcher to place the
analysis in time and to represent , when present, phenomena’s of “increasing returns”; which
makes the timing of the so called “small events” (Arthur 1994) or historical events determinant
for the final or evolutionary equilibria the system will reach.
By using a combination of institutional economics theory (mainly principal-agent) and
engineering design theory (mainly functional requirements engineering), our aim is to builda
systems dynamics model complemented by Role Playing Games that can capture the
institutional context and is able to indicate what contracting practices are likely to occur and
which ones are likely to succeed in view of the meeting the public values and demands. Several
international case studies will be used to build and validate our model. The research questions,
the theoretical background - mainly related to the evolution of institutions-, the research
approach proposed and the analysis realized making use a first conceptual simplified version of
this model will be presented in the paper from section 2 to 5. Section 6 present a discussion of
the remaining challenges and questions that need to be solved in order to arrive to the expected
simulation models. Finally in section 7, the most important conclusions of this paper re
presented.
2. Research Questions
The research project “Cross-national benchmark of Roads Infrastructures: Performance
Indicators and Innovative Contract Arrangements” aims to answer the following question: How
to achieve the results expected from innovative contracting - mainly improvements in efficiency
and innovation-, while keeping the room for strategic behaviour (of contractors) and the effects
of it at the minimum possible?
The main result expected from the research is thus an overview of lessons each of the
countries can draw on their own position and on how the other performs and to a evolutionary
and dynamic benchmarking framework that could serve as a tool for countries to evaluate if
elements from practices elsewhere can be incorporated in their own regulatory regimes,
contractual arrangements and practices at home
The most important sub questions are: a) What theories are relevant in order to analyze the
features of the different countries concerning innovative contracting policies in Road
Administration and the practices resulting thereof?, b) What are the contracting practices in the
different countries and how did they evolve?, d) What are the main similarities and differences
between the practices of these countries concerning the contracting and administration of roads?
The most important propositions are:
e The countries more successful in the use of innovative contracting practices were the
ones that implemented them in the more “coherent” way (proper timing, taking into
account synergy between different policies and overall coherency of the system), or
contrarily
e The ones that followed an “ad-hoc” approach, by applying each policy individually, not
paying explicitly attention to the coherency and synergy between policies, but rather
applying extra measures per each new policy implemented (i.e. the use of technical
measurements together with the use of Key Performance Indicators in order to reduce the
room for strategic behaviour).
e Changes have mainly as a reason the financial deficit experienced by the different road
agencies
e The innovative contracting practices increase the “room for strategic behaviour’, which
is used or not in dependence of institutional context
There is a tension between flexibility and public values
e The project portfolio of road agencies is becoming a more varied one. It is nota
complete change from traditional to new practices but is rather resulting in a
combination of many different contractual forms in the whole spectrum from traditional
to new; varying per project characteristics.
As can be seen in the propositions, central in the research is the issue of understanding the
evolution of the different national procurement strategies. It is for this reason that the following
section focus on reviewing the different economic theories that give support to our analysis of
institutional evolution. The contribution from other theories will be presented later in this paper.
3. Institutions and their evolution (C onceptual Background: Economic
Theory)
Infrastructures are complex multi-actor systems and therefore “institutions” play a key role
in them. The present research project focuses on the economic and technical angle conceming
innovative contracting practices in Road Administration and therefore the analysis of institutions
will also be focused on their influence on the economic behaviour of the agents involved in the
problem.
Though standard economics is highly abstract and virtually ignored and continues to ignore
the analysis of institutions; paying no attention to the influence of the institutional environment
on the behaviour of economic actors (Dequech 2003; Groenewegen 2005); there is a growing
interest in the study of institutions from other economic streams of though. Though there are
different schools of thought, all together this body of theory could be called Institutional
Economics. The differences between these schools of thought will be explored later in this
chapter.
Institutional Economics is expected to play a central role in the research project because of
two reasons. First, the research object itself, “innovative contracts” is a particular kind of
institution. A second reason is the comparative nature of the project. Only when transactions
costs - which are the costs that differ depending on the kind of institutions or governance
structures- are taken into account, it is meaningful to compare the institutional systems and/or
contracting structures of different countries. When transaction costs are neglected, the
differences in contracting practices and governance structures are also not expected to have
explanatory value, because in principle the performance and the costs of all of them should
eventually lead to the same optimal results. As expressed by Groenewegen (2005) one
shortcoming of standard economics is that it has no adequate explanation for the existence of
different organizational forms.
In the study of institutions we can distinguish different theoretical bodies and research
schools. First, we can make a distinction between the Old Institutional Economics (authors such
as Veblen, Mitchell, Commons, Ayres, Gordon and others) and the New Institutional Economics
(authors such as Williamsom, North, Langlois and others). Within these two, there are more
specific research schools like the so called Comparative Institutional Economics (CIA) with
writers as A oki and Evolutionary Economics or the Economics of Increasing Retums with
authors such as Arthur.
Rutherford (1994) points out that the dispute between the Old and the New Institutional
Economics centres on five main issues: a) Formalism vs. antiformalism, the role of formal
theoretical modeling as opposed to less formal methods, including historical and “literary
approaches”, b) Individualism vs. holism: the emphasis to be placed on individual behaviour
leading to social institutions as opposed to the effect of social institutions in moulding individual
behaviour, c) Rationality vs. rule following, the validity of rationalist explanations as opposed to
those that place limits on the applicability of rationalist conceptions, d) Evolution vs. design: the
extent to which institutions are the result of spontaneous or invisible-hand processes (resulting
from the maximization behaviour of each individual) as opposed to deliberate design, and e)
Efficiency vs. reform, the basis on which normative judgments can be made, and the appropriate
role of government intervention in the economy.
Given the purpose of this paper, we will explore in this section, only the difference
regarding the way these different research schools conceptualise evolution.
3.1. How do institutions change?
Central for understanding the roots of the differences between the contracting practices of
different countries is to analyze how these different systems have evolved. As Greif (1998)
points out, among the most fundamental questions of institutional economics are: “Why do
societies evolve along distinct institutional trajectories? And, why do societies often fail to adopt
the institutional structure of more successful ones?
Standard or Neoclassical economics is quite static on its point of view. For standard
economics just one single and most efficient equilibrium point is possible and therefore,
independent of the history or initial conditions of the system, all economical systems should
evolve until they reach the ideal and most efficient equilibrium point. This equilibrium point is
inspired on the Walrasian model and the A merican model is considered its nearest
approximation. The idea of a single equilibrium point is coherent with the assumption of
scarcity of resources and the law of decreasing returns.
New Institutional Economics (NIE) is richer in its analysis and through the inclusion of the
concept of transaction costs opens up the possibility of different arrangements or governance
structures being the most efficient dependent on asset specifity, frequency and uncertainty in the
transaction (Williamson 2000). Nevertheless, it keeps many of the static and formal
assumptions of neoclassical economics. Even though different arrangements are considered
plausible (E.g. hierarchy, market or hybrid), it keeps aiming at finding the most efficient
governance structure for a specific transaction. The main reason for this similarity in
conceptions is that NIE continues to assume the law of decreasing retums and a rational
maximizer individual - with bounded rationality in the worst of the cases-, not directly
influenced by exogenous or pre-existing institutions.
Though both Old and New institutionalism recognize that institutions may be deliberately
designed and enforced or may involve in unplanned or “spontaneous” processes, the authors in
NIE conceive the evolution of institutions more as a spontaneous and invisible-hand processes ,
unintended results of human action but not of human design. Meanwhile, the authors in the Old
Institutional Economics (OIE) question this view and opt for a more design oriented one.
Menger’s central questions give us a clear impression of this tension: How can it be that
institutions which serve the common welfare and are extremely significant for its development
come into being without common will directed toward establishing them?” ([1883] 1985 p.146)
Moreover, in what refers to the importance of history, OIE seems to give a higher weight to
it than NIE and becomes in this way seems to become more evolutionary. Veblen (1898) from
the OIE tradition conceptualizes the evolution of institutions as one of a “genetic account of an
unfolding process”. He aims at a treatment of institutional evolution as a process of “cumulative
causation”. For him, being this cumulative process more based on habituation to material
conditions and constraints -rule following- than on rationalistic calculation.
Rutherford (1994) highlights two important points rising from V eblen’s notion of
“cumulative causation”. The first is the statement that institutional evolution should be
conceived in a more wider sense than simply the adjustment to a series of exogenously given
shocks. In this sense, NIE seems to fall short because does not deal with the internal dynamic of
the system but only with the rational (or intendedly rational) responses of economic agents to
exogenously given changes in population, technology, trading opportunities , or ideology. The
second is that a discussion of cumulative causation involves a clear idea of path dependency.
This concept claims that history matters because the choices made earlier does constraints
the choices available in the future, what happens next depends critically on the present
conditions, which is also the outcome of the pre-existing situation. Therefore, it also claims that
small differences in initial conditions can make for widely differing outcomes.
These differences between NIE and OIE are only a general reading of the authors in these
schools. A deeper view of their works let us see that the dichotomies between traditions are also
present within authors in the same tradition. Therefore, it could be more helpful to explore the
theories of specific authors or the approaches of specific research schools instead of continuing
with the comparison of the NIE vs. the OIE view in general.
3.2. Comparative Institutional Analysis vrs. Evolutionary Economics
Two approaches will be discussed in this section, one coming from the Comparative
Institutional Analysis (CIA) research program, particularly based in the works of Aoki (1995;
2000; 2002) and one coming from the area of evolutionary economics, particularly based in the
works of Brian Arthur (1989; 1990; 1990; 1994) on increasing retumms and path dependence.
3.2.1. SINGLE EQUILIBRIUM VS. MULTIPLE EQUILIBRIA
Both approaches recognize the possibility for the evolution of multiple equilibria. A oki
states “I have suggested that the convergence of the various economic systems that exist
worldwide, and a complete convergence toward the Walrasian model in particular, would be
difficult because of the variance in historical conditions among economies and the need for
structural consistency between regulations and other institutions” (1995 p.5). Arthur makes also
reference to this when explaining what he found on the problems more interesting to him
(normally problems involving competition among objects whose “market success” was
cumulative or self-reinforcing): “There was typically more than one long-run equilibrium
outcome. The one arrived at was not predictable in advance; it tended to get locked in: it was not
necessarily the most efficient; and its “selection” tended to be subject to historical events” (1994
p. xiv).
As already shown in their statements, similarities but also differences are found in the way
they explain the processes from which these different equilibria result.
3.2.2. _ DESIGN OR SPONTANEOUS PROCESSES
Both authors seem to visualize these processes as spontaneous and invisible- hand, resulting
from the choices of individuals aiming at maximize their gains but with bounded rationality.
Bounded rationality means that individuals face constraints in their information processing
capabilities as well as on their maximization calculation capabilities. This change towards
bounded rationality is indeed what opens the possibility to more evolutionary explanations.
3.2.3. INSTITUTIONAL COMPLEMENTARITY AND PATH DEPENDENCE
They both are concemed with the “inertia” of the evolving system. Nevertheless A oki
explain this inertia and resistance to change in terms of Institutional Complementarity’, while
Arthur conceptualize this inertia or trend difficult to change as resulting from positive feedbacks
in the system for which the initial conditions and “small events” in the beginning matter. He
describes two characteristics for increasing return system; the first, inflexibililty, which refers to
the possibility of lock-ins and the second, nonergodicity. A bout this second property he says
“historical “small events” are not averaged away and “forgotten” by the dynamics. History may
decide the outcome”. It seems clear that the conceptualization of Arthur accentuates the
importance of history and “path dependence”.
| Institutional complementarity means that the operations of one institution are reinforced by the existence of other
institutions. The regulatory structure generally reinforces the system when it is internally consistent with other
systemic elements. Each element of the institutional structure is mutually reinforcing. Thus an institutional structure
will be resistant to environmental changes once it is formed. A ttempts to change one institutional element
independent of other elements will have only a limited effect Arthur, W. B. (1994). Increasing Returns and Path
Dependence in the Economy, The University of Michigan Press.
3.2.4. CONSTANT INERTIA OR TIME DEPENDENT INERTIA
They also differ in the way they conceptualize this inertia. For A oki this inertia of the
system is to some extent constant, while for Arthur the inertia of the system increases after
certain point, making a trend almost impossible to be changed. This difference can be seen in
the following statements. A oki says “Historical conditions extant (currently or actually existent)
at the outset of the transition constrain the possible outcomes of the reform process, but they do
not fully determine their later development” (1995). Arthur view instead says that “
heterogeneities, small indivisibilities, or change meetings - what he calls “small events”- become
magnified by positive feedbacks to “tip” they system into the actual outcome “selected” (1994
p.27).
3.2.5. | DECREASING VS. INCREASING RETURNS
As expected, the root of this difference can be found in their assumptions of decreasing
retums vs. increasing returns. A oki firmly states that “the reason for the evolution of multiple,
sub-optimal organizational conventions is not increasing returns as focused recently in
economics, but complementarity among strategic choices of agents (strategic complementarity)”
(1995). He defines strategic complementarity as the process of selecting the strategy that
corresponds to the prevailing social strategy. Meanwhile Arthur does believe that increasing
retums (or positive feedbacks) are the reason for the evolution of different and hardly or non-
predictable equilibria.
There are also differences in the way the resulting equilibria are conceptualized. Though
Aoki aims to go to a more evolutionary view that other authors in the NIE, he also try to stick to
the main assumptions of Neoclassical economics, as is the one that there is one resulting
equilibrium which is also the most efficient one. This tension can be clearly seen in his work. He
says “there is not absolute guarantee that the structure achieved will be the optimal one”(1995
p.11). He even seems to indicate that sub-optimal equilibria (reached through rational individual
decision) are possible, but he also states that really not efficient options are not possible
equilibria. Thus he stills consider that the resulting equilibria will be always efficient and at
worse, sub-optimal. Arthur is less confident in this respect and thinks that even though
individual choices are rational, there is no guarantee that the option selected (between two
technologies for example) is, from any long-term collective point of view, the better of the
available options.
An important question remains: Even though the inertia in the system, how does change
take place? The CIA approach continues to consider external factors as the cause of change.
Aoki says “Though the institutional structure is inertia-driven, the structure and environment of
the economic game underlying it will be affected by the development of production and
information processing technologies as well as by the globalization of financial transactions”
(1995).
In synthesis we could say that Arthur goes further than A oki in terms of evolutionary
explanations and challenges important assumptions of Neo-classical economic by assuming
increasing retums instead of Decreasing returns and the importance of Initial conditions.
Nevertheless in that way he also opens the possibility to explain the creation of multiple
equilibria in a more dynamic and to some extent more consistent way.
Moreover, thorough the emphasis he put on “small events” at the beginning that tip the
development in a particular direction and the inclusion of the time dimension, though the
formalization of the theory becomes certainly more difficult, it also becomes much more
dynamic.
Arthur himself says in the preface of this book “Increasing Retums and Path Dependence in
Economics”: “the increasing-retums world in economics is a world where dynamics, not static,
are natural; a world of evolution rather than equilibrium; a world of probability and chance
events” (1994 p.xx).
3.3. Making use of System Dynamics
The theory of Arthur keeps a close relationship to the thinking behind Systems Dynamics
theory. Both theories cite Prigogine and his advances in biology and source of inspiration of
their standpoint about the evolution of systems.
It is also important to underline that when the views of Aoki and Arthur are analyzed in
System Dynamics (SD) terms, some of the dichotomies seem to vanish. Both terms
“complementarity” and “reinforcement” from Aoki, and “increasing retums” from Arthur, make
reference to what SD calls “positive or reinforcing feedback loops”. Also when A oki refers to
institutional reform and advice successive systemic changes in place of a Big Bang approach, he
adds that “Rather, finding the source of strategically important and politically feasible
changes may open up opportunities to induce successive systemic changes by relaying on
“complementarity” between various institutions”. This view is very much in line with the idea
that SD propose of finding “leverage points” when changes in the system performance are
wanted.
For these two reasons, the similarity between SD and Arthur's approach and the potential of
SD as a tool to find a middle ground between these two theories, we think that if the evolution
of the contracting practices is to be modelled, System Dynamic methodology offers a promising
potential.
It is now clear why System Dynamics should play an important role in the research project
and why it has been chosen as the right modelling and analysis technique. Now we can proceed
to the following section, where the research design is presented.
4. Research approach proposed
In order to answer the research question and to validate the propositions already mentioned,
the research has been designed as a multiple case study research with embedded units of
analysis. The main unit will be the national procurement strategy (or the strategy of the road
authority in charge of national roads and highways) with all its components (i.e. contractor
selection methods, project delivery, project finance and performance contracting between
others). The embedded unit will be projects, preferably Public Private Partnerships (PPP)
projects in which these new contracting practices are being used.
Possible case studies will be one Scandinavian case (Finland and/or Norway), one Hispanic
(Spain and/or Chile) and one or two Anglo-Saxon cases (England and Australia). The methods to
be used in order to draw conclusions from the evidence gathered in the case studies will be
mainly System Dynamics and Role-Playing Games. These two methods will be used in a
complementary fashion. This means that Role Play Simulations will be carried out taking into
account values given by the SD model, and at the same time, the results from the Role-Playing
Game Simulation will be feed into the SD model. The Role-Playing simulations will particularly
deal with the soft variables in the model, such as “Strategic behaviour”, which are the product of
human random decisions.
The steps realized and still to be done, in order to arrive to these simulation models are the
following:
1) System Analysis. A first set of causal diagrams has been built as result of a meta-analysis. A
significant number of articles in the field has been reviewed and this together with interviews
with experts has resulted in a first draft causal and generic model which includes a)Institutional
and evolutionary economics concepts, b) Project delivery methods, c) Project financing methods
and d)Performance contracting. The results from this first step are presented in the following
section.
2) Conceptual model application to the different case studies. This will be done in two steps.
First the generic causal model will be applied to each country, reflecting the different stages in
the reform process and considering as main policy only contracting practices that count for 30%
or more of the projects. Second, the model will be applied to the embedded unities or specific
innovative projects. This will give us a view at the future and we could use the parameters of
this model to run a prediction on evolutionary stages; as well as new mechanisms in the system
will be discovered in comparison to the initial system studied in step 1. It is in this model or
specific case that more attention will be paid to the use of Key Performance Indicators and
design space or freedom. These aspects are very difficult to study for the whole country.
3) A Role-Playing Game simulation will be developed also per case study. This will contain the
general plot of actors, formal and informal institutions linking actors and constraining their
behaviour and a clear logic from actors decisions to overall outcomes, which will be the ones
used as input of the SD model.
4) Development of a SD simulation model. Parallel to the development of a Role-Playing game
simulation, the development of SD simulations model has also to take place. It is important to
mention here that we are aiming not at a single SD model that shows all the different phases the
country has gone through because with each change in policy the structure of the system itself
also changes and therefore a single model is almost impossible to be done. Instead we are
aiming at different SD model that show; a)The system before the reforms (with the specific
values of each variable per country); b) Country specific models that will help us to predict the
possible effects of the application of new policies (the values of it will come from the data of the
innovative project) and finally, after the conceptual models have been validated and applied to
each case study, will be possible to generalize and build b) 3 or 4 models that show the different
combinations of policies or strategies that are now in place in different countries . These models
could be eventually used by different policy makers by making them specific per country when
placing the national values per parameter and perhaps varying the strength of the causal
relationships
5. System Analysis
In order to answer the questions previously mentioned and after a preliminary research -
including interviews and literature review- a system analysis was carried out and a first
conceptual model of the procurement system was built. The model consists of a set of system
diagrams that explain the evolution of the four main trends in road contracting. In this paper,
only the two final diagrams will be shown. The complete set of diagrams was presented in
(Altamirano and Haraldsson 2005) and (Altamirano and Herder 2006). A brief version of this
dynamic view will be presented in section 5.2.
5.1. Principal Agent Theory and Engineering Design Theory
Before presenting system diagrams with the analysis of the evolution of the different
practices, we will present here the most important causal relationships in the model. These
relationships together with their theoretical basis are presented in Table 1. The relationships not
mentioned in this table come from the literature of road administration or from interviews with
experts.
Causal Theory | Theoretical explanation
Relationship
Output/effort based Design The more operation and detailed are formulated the client requirements (for
criteria >(-) Design | Theory tendering) the smaller the design space left to the contractors bidding
space
Flexibility > (+) Design The more design space given to contractor, the more opportunities he has to
Innovation Theory innovate (in the design solutions he will propose)
In-house expertise Agency | Information asymmetry occurs when one party to a transaction has more or
>(-) Information Theory better information than the other party and is therefore equivalent to what
Asymmetry Agency Theory calls “private information” (Lambert 2001) of the agent. If
the government agency has less expertise and technical know-how, the
“private information” of the agent increases.
Information Agency | Taking into account that the agent has an incentive to misreport the “signal”
Asymmetry >(+) Theory he saw in order to receive a more “favourable” compensation contract, if
Room for there is more information asymmetry and he has more “private information”
opportunistic he will have more opportunities to misreport without being noticed.
behaviour
Outcome/Final Agency | This is the result of various theoretical propositions: a) The sensitivity of the
Quality Criteria > Theory signal (performance measure) measures how much the expected value of the
(+) Information signal moves in response to a change in the agent's effort, b)Effort is costly
asymmetry >(+) to the agent and therefore he aims to minimize it,
Room for strategic Assuming that the higher the level of performance indicator used, the more
behaviour noise is introduced in the signal and this become more driven by other
factors (i.e. external factor, etc) than agent’s effort; then information about
agent's effort become private information. Therefore outcome/final quality
criteria are expected to give more opportunities to the agent to maximise his
utility without increasing the real outcome and own effort level.
Combined/Integrated | Agency | Given that not all performance measures are equally sensitive to a particular
contracts > (+) Theory action, some can be more easily manipulated than others. When the agent is
Room for strategic responsible for multiple tasks, he can vary how much attention he spends on
behaviour one task versus another and in this way manipulate the compensation he
receives.
Long term of contract | Agency | Multi-period models make the management of principal-agent relationship
> (+) Room for Theory even more complex. As is the case when the agent is responsible for multiple
strategic behaviour tasks, the private information of the agent becomes larger. When he can
decide on investment levels for different periods, he can also distribute
investments in the most productive way for him, which is not necessarily the
one more convenient for the infrastructure condition and future performance.
Table 1. List of most important causal relationships. Source: (Altamirano and Herder 2006)
5.2. The conceptualization of the evolution process
The dynamic view that resulted from the system analysis carried out can be formulated in
brief as follows:
The environment of the public sector has changed significantly in the last decades. A new
paradigm has emerged in public administration causing a global trend of liberalization and
10
privatisation of public utilities, including road infrastructure. This trend together with the
stagnation of national economies has created a funding problem.
In reaction to this funding problem some countries have started to experiment with
indirectly financed projects (financed by private parties, so as concessions and toll roads).
Besides the problem in funding, another effect of this process of liberalization and privatisation,
has been the loss of in-house expertise, which weakens the effectiveness of direct supervision
methods.
Thus, once more a solution to this new problem is needed. Key Performance Indicators
(KPI) are beginning to be incorporated in contracts. These indicators are more outcome that
output or effort oriented (see Figure 1). The positive effects of this measures is that small public
agencies can manage the road network efficiently focusing only on the key strategic issues; and
that more flexibility and therefore design space is granted to contractors; increasing the room or
the probabilities for innovation (i.e. materials and processes).
Direct financed —|ndirect financed Segmented Qutcome/ Final — Short term of
Projects projects contracts (DBB) Quality criteria contract
tad \eeegeeoeesn wannnamanege tes iedaskan aes
oa Private Risk of
q ile Resources ‘av: Investing in | D lity.of
; available for Ri q ~ Quality of
Paradigm: Global Trend { need ch t_-¥ Re HEED bill {Technologies ] Roads
of Liberalization and! pe Soneen| | Fle ibiity ib
Privatization ~Se)_pupie =| Funding a contootons H
'- | Resources | - |_9aP _1 a ,
= (NEE for y+ - ——— iI H
Steady National T {information| + AooM pene 1 + yinnovation
Foonomies: Inhouse ] ~ [assymetry [ [behaviour ~ | Monitoring i
t expertise i
Figure 1 System Diagram of the Second Trend in Procurement Strategies
Nevertheless, this solution also creates new problems. The match between responsibility
and influence has become in risk creating more room for strategic behaviour. A focus on
outcomes/results (i.e. number of accidents in the road) could mean that contractors are being
held responsible for things out of their area of control or influence.
Indirect i |
oun a
proj ects
- outsorced
Public \
“"e
In-house technical
ae ofthe 4 expertise
Road Agency
Figure 2 Detailed view of reinforcing loops in the system
It is important to highlight that even though these have not been pictures in the system
diagrams - for the sake of simplicity- many reinforcing loops are taking place in this reform
process. An example is presented in Figure 2. As it can be seen, there are two reinforcing loops
in the system that together accelerate the reform process. The first can be explained in the
following way: the less the public resources available for the road agencies -as product of the
11
new Public Administration paradigm and economic stagnation- the more the need and the higher
the number of projects that will be probably financed by private parties. At the same time, the
more the indirectly financed projects, the more success stories for PPP projects will arise and the
preference towards this kind of projects will be reinforced, doing less and less acceptable the use
of public funds for road infrastructures projects and therefore actually causing a reduction in the
public resources available for the road agencies.
This reduction in the public resources assigned to the road agencies cause also a downsizing
of the agency and therefore the loss of personnel with technical expertise. The less the technical
expertise in-house, the higher the number of new projects that will be outsourced, which at the
same time makes technical expertise less necessary and therefore accelerates the process of
emigration of personal with this expertise towards the private companies. Finally, it also
happens that the more indirectly financed projects, the more the number of projects that will be
outsourced, because usually PPP projects involves and require a private company not only for
the financing but also for the realization of it. In this way the result is two reinforcing loops, out
of which the first accelerates the second even further, through its two factors, public resources
available and indirect financed projects.
The evolution process continues in this same way, a new policy or solution is implemented
and with this a new problem arises, making new policies necessary. Until we finally arrive to a
scenario with four new policies, all part of the national procurement strategy; 1) Indirect
financed projects, 2) Combined contracts, 3) Outcome/Final Quality criteria and 4) Long term
contratcts (see Figure 3).
Direct financed Indirect financed Combined Outcome/ Final —_Long term
projects projects contracts Quality criteria contract
Private Life Cycle| a
Resources Costs \ 7 Risk of
available for ral awe. Investing in
Match} Technologies
‘Quality of
Roads
a
@)” Public Jf] Funding |” IResponsabill
Resources gap
needed for fy/linfivence
+
EN
Flexibility
or
contractors
Paradigm: Global Trend
of Liberalization and
Privatization
“| In-house.
hsbc 5
| *>| expertise <<
t | Resources z .
| Available for \ty _, |Roomior! _lettectiveness! | [nnovation
Steady National ___—r*| RA’ Informationt#Lbehaviour| ~ of |
Economies Sey Monitoring | |
Figure 3 System Diagram of the Fourth Trend in Procurement Strategies
This first analytical exercise already shows that the reform process in the Road
Administration sector has been a gradual process of multiple instable equilibria, and initiated
originally by changes in external factors and paradigms in public administration. Changes in the
environment created a new set of problems. Solution to these problems, created new ones and
successively in that in the way, the countries have been forced to implement these different
trends, though sometimes in different orders.
12
6. Preliminary Results and Discussion
The first system analysis of the evolution of the different trends allows us conclude that.
In general, one needs to be careful about the results expected from the application of these
practices. The attractive efficiency wins and savings from some countries seem to have been the
result of proper timing and application order of the different strategies, probably combined with
favourable extemal factors and extra policies that counter balance the possible perverse effects
of some of these policies. The system analysis also seems to indicate that the different trends are
interdependent and to some extent some are the logical reaction to problems caused by the
“solution” given in the previous trend. Synergy is expected for example towards achieving
innovation. If policies increasing the flexibility or design space or contractors are implemented
together with policies that reduce the risk for the contractor to invest in new technologies, their
combined effect in innovation is expected to be larger than when applied individually.
Finally, the tension between technical aspects and public values was clearly captured in the
system diagrams: increasing flexibility has had an impact in the room for strategic behaviour
(Altamirano and Herder 2006). Flexibility has been increased aiming at improving innovation,
the service levels and therefore aiming at achieving public values more efficiently. Nevertheless,
the policies implemented to increase flexibility seem also to increase the room for strategic
behaviour, which -when used by the contractors- can translate in quality decline and therefore
could actually threat the fulfilment of public values in the long term.
Nevertheless, in order to draw more concrete conclusions about the pro’s and con’s of
applying these new contracting practices, about the preconditions for success and of course to
answer the overarching research question of this study: How to achieve the results expected
from innovative contracting - mainly improvements in efficiency and innovation-, while keeping
the room for strategic behaviour (of contractors) and the effects of it at the minimum possible?
We need to go further and after applying these conceptual models to specific case studies, also
translate them in concrete System Dynamic simulation models that could predict the results of
different combinations of policies, the possible conflicts between practices and show the best
ways to solve the trade-off between flexibility and public values.
This translation poses big challenges and important questions, some of these are:
a) Should we build one single generic model with many decision rules or indeed the
proposed set of different models. How can we deal with the changes in the structure of
the system itself that result from the implementation of new policies?
b) Can we model the problem at this level of aggregation, with variables such as in-house
expertise and room for strategic behaviour; and still obtain significant accurate values for
the results in performance of these new practices?. Or do we need to model the
underlying physical flows and stocks?
c) Do we need to model the infrastructure itself or can we model only the operation of it?
d) How do we model “room for strategic behaviour’? Can we model it making use of
System Dynamics of it is really necessary to use Role-Playing Games?
e) Is the combination of System Dynamics with Role-Playing Simulation a
methodologically valid and promising one?
The discussion of this paper with experts and other researchers using System Dynamics for
similar problems aim to reflect on these questions and find good ways to solve these modeling
dilemmas.
13
7. Conclusions
After presenting an overview of the research questions being proposed, the conceptual
background that will be used to study the evolution of the procurement strategies, the research
approach being proposed and the preliminary results of the system analysis carried out; it can be
concluded that System Dynamics is indeed a methodology with great potential for the modelling
of evolutionary processes where multiple equilibria are possible, where historical events and
initial conditions do make a difference and where the time domain needs to be included.
Nevertheless, the use of System Dynamics for explaining these evolutionary processes -
given the nature of many of these problems in which institutional contexts and many
stakeholders play an important role- poses still important challenges regarding the way these
“soft” variables -like strategic behaviour- resulting from the choices of diverse economic agents
can be modelled and how the structural changes resulting from new policies, can be integrated
ina single deterministic model. A combination of SD models with Role-Playing Games is
proposed in order to overcome the first difficulty but results from this combination are still to be
seen.
8. References
Altamirano, M. A. and H. V. Haraldsson (2005). Resolviendo juntos los problemas de las
Infraestructuras de la Proxima Generacion. III Congreso Latinoamericano de Dinaamica
de Sistemas, Cartagena, Colombia.
Altamirano, M. A. and P. M. Herder (2006). Systems Dynamics modeling for road contracting
(under review). Transport Research Arena, Europe 2006, Goghenburg.
Aoki, M. (1995). Information, Corporate Governance and Institutional Diversity.
Competitivenes in Japan, the USA and the Transitional Economies. New Y ork, Oxford
University Press.
Aoki, M. (2000). Institutional Evolution as Punctuated Equilibria. Institutions, Contracts and
Organizations: Perspectives from New Institutional Economics. C. Menard. London,
Edward Elgar.
Aoki, M. (2002). "Toward a Comparative Institutional Analysis." Journal of Institutional and
Theoretical Economics JITE 158(4): 735-737.
Arthur, W. B. (1989). "Competing Technologies, Increasing Returns and Lock-in by Historical
Events." Economic Journal 99(March): 116-131.
Arthur, W. B. (1990). "Positive Feedbacks in the Economy." Scientific A merican(February): 92-
99.
Arthur, W. B. (1990). "Silicon Valley Locational Clusters: When do Increasing Retums Imply
Monopoly?" Mathematical Social Sciences 19(235-251).
Arthur, W. B. (1994). Increasing Returns and Path Dependence in the Economy, The University
of Michigan Press.
Cervera, A. and R. E. Minchin (2003). The Change to End-result Specifications: Where Are We
Now? TRB 2003 Annual Meeting CD-ROM.
Cox, D. O., K. R. Molenaar, et al. (2002). Contract Administration: Technology and Practice in
Europe. O. o. I. Programs, O. o. Policy, F. H. Administration and U. S. D. 0.
Transportation, American Trade Initiatives.
Dequech, D. (2003). The New Institutional Economics and the theory of behaviour under
uncertainty. EAEPE XV Conference: The Information Society - Understanding Its
Institutions Interdisciplinarily, Maastricht, the Netherlands.
14
Greif, A. (1998). "Historical and Comparative Institutional Analysis." American Economic
Review, Papers and Proceedings: 88-84.
Groenewegen, J. (2005). Behaviour of economic actors influenced by formal and informal
institutions. TPM Quarterly. Faculty of Technology, Policy and Management, TUDelft.
II: 7.
Herk, S. v., P. M. Herder, et al. (2004). Opportunities and Threats for Granting More Design
Space to Road Contractors*. IEEE International Conference on Systems, Man and
Cybemetics.
Lambert, R. A. (2001). "Contracting Theory and accounting." Journal of Accounting and
Economics 32(2001): 3-87.
Manley, K. and S. McFallan (2003). Innovation A doption Behaviour in the Construction Sector:
The Case of the Queensland Road Industry. 2nd International Conference on Innovation
in Architecture, Engineering and Construction,, United Kingdom.
Menger, C. ([1883] 1985). Investigations into the Method of the Social Sciences with Special
Reference to Economics. New Y ork, New Y ork University Press.
Miller, J. B. (2000). Principles of Public and Private Infrastructure Delivery. London, Kluwer
Academic.
Pakkala, P. (2002). “Innovative Project Delivery Methods for Infrastructure -An International
Perspective."
Rutherford, M. (1994). Institutions in Economics. The Old and the new institutionalism.
Cambridge, UK, Cambridge University Press.
Veblen, T. (1898). "Why is Economics Not an Evolutionary Science?" Quarterly Journal of
Economics 12: 373-397.
Williamson, O. E. (2000). "The New Institutional Economics: Taking Stock, Looking A head."
Journal of Economic Literature XXX VIII(September): 595-613.
15
