Confirmations and ContradictionsThe Leontief Paradox, Reco.docx

Confirmations and Contradictions
The Leontief Paradox, Reconsidered:
Correction
Iraj Heravi
University of California, Los Arigetes
Leontief (1954) incorrectly inferred that the United States is
revealed
by trade to be relatively more abundant in labor than in capital
from
the fact that the capital per man in imports exceeds the capital
per
man in exports. The counterexample in Leamer (1980) is in
error,
however, since it uses commodity prices at which the first
commodity
is not economically produced. This is easily remedied by setting
the
price vector to (2, 1, 1) instead of (1, 1, 1). Then, by the series
of
calculations outlined in Leamer (1980), it is possible to
compute the
ratio of capital per man in exports to capital per man in imports
to be
0.97, even though capital is the relatively abundant factor.
Incidentally, the error in the choice of prices is revealed by the
computation of the returns to factors, which for the original
numbers
contains a negative return to capital.

References
Heravi, Iraj. "Determination of Patterns of Trade." Ph.D.
dissertation, Univ.
California at Los Angeles, 1984.
Leamer, Edward E. "The Leontief Paradox,
Reconsidered."/.P.£:. 88 (June
1980): 495-503.
Leontief, Wassily W. "Domestic Production and Foreign Trade:
The Ameri-
can Capital Position Re-examined." Econ. Internazionate 7
(February 1954):
9—38. Reprinted in Readings in International Economics,
edited bv Richard E.
Caves and Harry G.Johnson. Homewood, III.: Irwin (for
American Econ.
Assoc), 1968.
Postdoctoral scholar. Graduate School of Management, UCLA. I
am indebted to
Edward Leamer for his advice and encouragement. See my
dissertation (1984) for
further work.
[Jounuil (i/ Piiliiiral Ecimomy. 19R6, vol. 94, no. 5]
© 198(3 by l h e L nivfisuy of Chicago. All rights reserved.
0022-3808/86/9405-0010$01.50
1 12O

#1. DUE Thursday 1/ 22 /2015 NOON…. Continue working
with the company you originally chose from the Fortune’s 500
Best List ( Johnson & Johnson)
http://archive.fortune.com/magazines/fortune/mostadmired/2006
/best_worst/worst8.html
Detail Directions (See Below) – CLASSROOM FORUM
DISCUSSION
1. You have been working with each environment
individually; now it's time to integrate and synthesize.
2. Perform research (minimum of 3 sources in APA format).
3. Analyze each environment collectively by incorporating
theories learned.
4. How are the individual environments working together for
your chosen company?
5. How is each working with the other environments?
6. Is an environment causing limitations on another? If so,
which environments are involved and what are the limitations?
7. What could you change to create synergy? Your answer
cannot be that you wouldn’t change a thing. You must
incorporate critical thinking.
8. Minimum 6 complete paragraphs; a paragraph is a
minimum of 100 words.
*****Use this as one of the References: Use at least one of
the sources provided below in “Reading / Resources: “.
#2 . DUE Saturday 1/ 24 /2015- NOON . Continue working
with your chosen company from Fortune’s 500 Worst List (
Chosen Company is “Google”)
http://archive.fortune.com/magazines/fortune/mostadmired/2006
/best_worst/worst8.html
Detail Directions (See Below) – FORMAL PAPER
Assignment Instructions
Continue working with your chosen least admired company

(Google) for the assignments and complete the following:
1. Identify the company’s domestic environment and discuss
how the government regulations affect its domestic environment
it must operate in.
2. Identify a global environment for the company and
discuss how the government regulations affect its global
environment it must operate in.
3. Identify the hard and soft technology and interpret the
characteristics the company should have/use to be successful in
its domestic and global environment.
4. Identify the political-legal barriers for the company in
both the domestic and global environments. Use business
theory/theorists to illustrate how the company can operate
successfully in its domestic and global environment.
5. Identify, compare, and contrast sociocultural factors of
the domestic and global environments of the company.
6. Compare and contrast two economic theories for both the
domestic and global environments of the company.
7. Develop a strategy of success based on your evaluation of
steps 1-6 by assessing what you’ve learned through your
research and readings and compare what the company has been

doing to what you recommend they should be doing. You cannot
state that you would not change anything.
8. You must incorporate critical thinking (see resources).
9. Research requirement: minimum 5 sources PLUS the text
and Geert Hofstede.
10. Page requirement: 7 pages in APA format (does not count
cover page and reference page).
Note: ALL Assignments are submitted to turnitin.com and
checked for originality.
****Use this as one of the References: Use at least one of the
sources provided below in “Reading / Resources: “.
ADDITIONAL NOTES / INFORMATION:
This is it! Here you will finish integrating the environments of
your chosen company. At this point, you should have all
elements identified in your previous writings ( forums and paper
assignments). Is there anything you would change at this point?
Do you think you would be successful? Why or why not? The
assignment this week will be a final simulation project for your
chosen company. Using your company you’ve been working
with, recognize all the environments you’ve studied – domestic,
global, technological, political/legal, sociocultural, and
economic. All environments must be included. Once you have
each identified, it’s time to analyze and synthesize. What is
each individual environment like? How are they working? What

issues do you see? What would you change and why? It won't be
acceptable to state you wouldn't change anything. Really apply
what you've learned from your research.
Now, take a look at all environments together, as whole and not
individual pieces. Is each working together? Is an environment
causing limitations on another? If so, which environments are
involved and what are the limitations? What could you change
to create synergy?
All Sources must be “Scholarly Article or Book”. A scholarly
article or book generally is based on original research or
experimentation. It is written by a researcher or expert in the
field who is often affiliated with a college or university. Most
scholarly writing includes footnotes and/or a bibliography and
may include graphs or charts as illustrations as opposed to
glossy pictures. In addition, articles that appear in scholarly
journals or book that are published by academic presses, are
subject to a peer-review process, which means that other
"experts" or specialist in the field evaluate the quality and
originality of the research as precondition of publication. The
peer-review (as opposed to editorial review) process is also one
thing that sets scholarly journals apart from journals that may
otherwise seem quite similar. Journals such as Foreign
Affairs, for instance, are generally not considers "scholarly
journals," because many of the articles are solicited by the
magazine's editors; in addition many of the articles are written
by policy-makers who may be expressing an informed view, but
whose article may not be based on original research.Scholarly
research is typically published by a academic association or a
university/academic press. In international relations and
comparative politics, representative scholarly journals
include Asian Survey, Comparative Politics, International
Organization, International Security Studies, Journal of
Comparative Politics, Journal of Democracy, and World
Politics.
Reading / Resources:

You are not limited to the below:
**Review all previous resources provided.
Lin, H., Fan Y., & Newman S. (2009). Manufacturing process
analysis with support of workflow modelling and
simulation (p1773-1790. 18p. 6)
Závadský, J., & Lukáš, T. (2010). Simulation and its purpose in
implementing of business process management. Advances in
Management, 3(3), 9-12
International Journal of Production Research
Vol. 47, No. 7, 1 April 2009, 1773–1790
Manufacturing process analysis with support of workflow
modelling and simulation
Huiping Lin
a,*, Yushun Fan
a
and Stephen T. Newman
b
a
Department of Automation, Tsinghua University, Beijing,
100084, P.R., China;
b
Department of Mechanical
Engineering, University of Bath, Bath, BA2 7AY, UK

(Received 15 December 2005; final version received 14 August
2007)
Process analysis is recognized as a major stage in business
process reengineering
that has developed over the last two decades. Manufacturing
process
analysis (MPA) is defined as performance analysis of the
production process.
A manufacturing process analysis framework is outlined with
emphasis on linking
a company’s strategy to operational process. Two issues,
namely process
modelling and simulation based analysis, are investigated. A
compound workflow
model (CWM) is proposed to provide graphic presentation of
the production
process that can be easily understood. Also it can be used
directly by simulation
to study the impacts of scheduling policy and analyse the
process performance.
A two-stage simulation analysis method is provided to
quantitatively and
efficiently define cause-and-effect relations to identify drivers
for improvement.
The manufacturing environment, PSC (production planning,
scheduling and
control) factors and the process structure are three main
concerns considered in
the simulation. An example is discussed in the final part of the
paper.
Keywords: manufacturing process; performance analysis;
modelling; simulation

1. Introduction
With increasing market competition, more and more companies
recognize the importance
of improving their competitive capability. Over the last two
decades, business process
reengineering (BPR) has been carried out by academic
researchers and industrial
companies to improve performance. It concerns the fundamental
rethinking and radical
redesign of a business process to obtain dramatic and sustained
improvements in quality,
cost, service, lead time and innovation (Hammer and Champy
1993). Process analysis is
recognized as a highly important stage of any BPR project. It
understands the way the
process is done, finds problems and the gaps between current
performance and expected
targets, and identifies changes needed for improvement.
Manufacturing process analysis, which is defined as modelling
and performance analysis
of a production process, is essential for manufacturing
companies to improve their market
competition. Manufacturing is a complicated system that
involves sets of tasks, materials,
resources (including human resources, facilities and software),
products, and information.
Askin and Standridge (1993) divided the manufacturing system
into five interrelated
*Corresponding author. Email: [email protected]
ISSN 0020–7543 print/ISSN 1366–588X online
� 2009 Taylor & Francis
DOI: 10.1080/00207540701644151

http://www.informaworld.com
functions: product design; process planning; production
operations; material flow/facilities
layout; and production planning and control. Product design is
responsible for constructing
the description of the products. Process planning entails the
specification of the sequence of
operations required for product production. Production
operations refer to those
fabrication or assembly actions such as drilling a hole or
inserting a raw material into the
workstation. Facilities layout is concerned with the physical
placing of production process.
Production planning, scheduling and control (PSC) is an
important task within
manufacturing. It generates long/medium term production plans
and disaggregates them
to obtain short term schedules about the tasks’ sequence on the
machine.
The MPA discussed in this paper is mainly concerned with
production process
reengineering. It uses a top down method to associate the
company’s strategy with the
process’s target. By process analysis, it finds out the initiatives
to implement the target.
MPA is not the same as the production scheduling and control
problem. MPA optimizes
the production from the point of view of process structure
redesign. Most PSC methods
optimize the measurements such as time, and cost with the
existing resource or process

constraints. (Readers are referred to a series of review papers
about PSC namely Gupta
et al. 1991, MacCarthy and Liu 1993, Sox et al. 1999, Potts and
Kovalyov 2000, Tan and
Khoshnevis 2000.) MPA can result in adjustments of process
constraints. Process
improvement solutions possibly contain reconstruction of the
process or the organization,
for example, outsourcing some activities, adding a new
resource, reducing an activity’s
processing time by new technology, etc. It considers the
influence of PSC because the
appropriate planning and scheduling methods can also improve
the process performance.
Modelling and simulation based analysis of MPA are discussed
in this paper. Usually
a process model is established to understand the structure of
manufacturing, reveal
relationships among machines, tasks and materials, and learn
how the process works.
In the literature, mathematical models and graphic models are
two major methods to
describe the manufacturing process. For example, fuzzy mixed
integer programming
has provided the ability to model and analyse the manufacturing
cell formation problem
(Tsai et al. 1997). Yun and Gen (2002) used constraint
programming to establish a
pre-emptive and non pre-emptive scheduling model. Cooke et
al. (2004) developed mixed
integer programming formulations with the assumption of a
production precedence
sequence to study the economic lot scheduling problem.
Mathematical models are often
used for production scheduling purposes. Although

mathematical models are convenient
for computing, they do not offer an intuitive understanding to
their users. Petri nets are
a popular graphic method for modelling, scheduling and
analysing the manufacturing
process (Shih and Sekiguchi 1991, Lee and DiCesare 1994,
Proth and Minis 1995, Xiong
et al. 1996, Zhou and Venkatesh 1999). Shih and Sekiguchi
(1991) and Lee and DiCesare
(1994) used a Petri net with heuristic search for FMS (flexible
manufacturing systems)
scheduling. Xiong et al. (1996) and Zhou and Venkatesh (1999)
studied the application of
Petri nets in modelling, simulation and control of flexible
manufacturing systems. The use
of Petri nets has the advantage in describing complicated
process constraints. However,
the major problem is that its state space for searching grows
dramatically with system
complexity (Zhou and Venkatesh 1999). It still lacks a
comprehensive model that has all
following features:
. The capability to describe complicated relations among tasks,
machines, and
production routings;
. Graphic presentation so that the model can be easily
established and used;
1774 H. Lin et al.
. To fulfil requirements of MPA that is concerned with the PSC
and process

planning.
Thus, a workflow modelling and simulation based method is
proposed to solve the
problem in manufacturing process analysis. The contributions of
the paper are:
. Providing a manufacturing analysis framework that emphasizes
on associating
enterprise strategy to process.
. Providing a graphic presentation of a manufacturing process,
not only showing the
structure of a process, but also containing enough information
for dynamic
performance analysis.
. Proposing a structured two-step sensibility analysis to
quantitatively and efficiently
identify cause-and-effect relations to implement the strategy.
The paper is organized as follows: initially, an analysis
framework is introduced. How
production process modelling and simulation are used in MPA
is described. Then a
compound workflow model (CWM) is proposed to describe the
production process.
A two-step simulation method is explained in detail. The final
part of the paper provides
a case study which shows the effectiveness of the proposed
method.
2. Manufacturing process analysis framework
MPA should have a strategic level of thinking about the
objective and target of the

manufacturing process. It encourages people to be more creative
and more focused on
process improvement. More and more people have realized
strategic thinking in a project
will help to identify the desired performance and make
improvements more focused and
purposeful. For example, King (1994) pointed out the lack of
strategy accounted for many
failures in process improvement projects. Kaplan and Norton
(1992, 1996) proposed
a balanced scorecard (BSC) – namely an effective method for
strategy management.
In Kaplan and Norton’s method, four steps are involved: first, a
company’s strategy is
translated into process’ objectives. Second, key indicators are
selected to measure process
performance. Then, targets are associated with indicators at
operational level. Finally,
initiatives to pursue targets are identified and tasks will be
completed. In Kaplan and
Norton’s method, a cause-and-effect analysis served as a core
part in linking strategy to
process. The cause-and-effect relations are a set of hypotheses
about how to achieve
important objectives. Usually, it can be expressed as a sequence
of if-then statements: If we
do actions, then we may be able to achieve objectives.
Although BSC provides a framework for connecting a
company’s strategy to a process,
it does not specify how the connection can be built. Only when
people have a very good
understanding about a process’s structure and performance, can
they know which factors
account for what results. For a complicated manufacturing
process, although some

historical data and employee experience exists, it is very
difficult to quantitatively define
cause-and-effect relations. On the other hand, in Kaplan and
Norton’s straight down four
step method, there was no communication between the decision
making level that set
targets and the operational level that has details about
manufacturing. If the target is too
stressful, it will be expensive or even impossible to realize. It
will not benefit process
improvement projects until the target is appropriately set. Thus,
a target evaluation and
a bridge between the decision making level and the operational
level are needed.
International Journal of Production Research 1775
Moreover, as performance improvement solutions may have
structural change to the
process, feasibility verification of the solutions is required
before it is put into
implementation to avoid unnecessary loss.
Thus, a new manufacturing process analysis framework is
provided (see Figure 1).
An adapted workflow model and its simulation will be used to
clarify the manufacturing
process, understand how the process operates and access
process improvement solutions.
A feedback channel is introduced to help in setting reasonable
targets.
After the company’s strategy is clarified, the following steps
are taken to link the

strategy to the processes.
Step 1: Translating the company’s strategy to processes’
objective.
Connecting strategy directly to the process helps to break
through the boundary of
organizations and focus on process-oriented analysis. At this
stage, the process’s objective
is determined according to the company’s strategy. The
manufacturing process model is
needed for two reasons. Firstly, the manufacturing process
model defines the structure and
constraints for the manufacturing processes. It provides a
platform for employees and
project teams to understand the manufacturing process and
communicate with each other.
Secondly, defining the boundary of the processes also helps the
manager to determine the
objective of the process.
Step 2: Identifying key process measurements according to the
process objective.
Measures for manufacturing performance can to be catalogued
in terms of business,
operational and customer perspectives (Challis et al. 2002). As
performance of the
manufacturing process is our main concern, operational
performance is our major focus.
The following performance measures are widely used in
manufacturing: production
throughput, lead time, reliability, cost, capacity, resource
utilization and product quality.
Readers can refer to Altiok (1996) and Viswanadham (1999) for
definitions of some typical
measurements in manufacturing environments.

Step 3: Associating a rough target with selected indicators.
Figure 1. Manufacturing process analysis framework.
1776 H. Lin et al.
At this point, a rough target is set by estimation of cause-and-
effect relations with historic
data and employees’ experiences.
Step 4: Finding initiatives to complete the process improvement
task and verifying the
feasibility of the target.
Improving the production throughput and then maximizing the
benefit is the most popular
objects for production improvement. There are two major ways
to achieve this. The first
method is to take the existing production capacity as given and
optimize the organization’s
performance within the capacity. According to Goldratt and
Cox’s (1992) theory of
constraints, the manager can increase throughput by relieving
the bottleneck factor of the
production. They often do this by choosing an effective
scheduling method to reduce

the downtime on the constraining resources or buffering this
resource; or reengineering the
production to reduce the demand on constraining factors. The
second method is following
the long term operations strategy to increase capacity. The
process improvement can be
catalogued in four dimensions such as process level, activity
level, resource level and
management level. The typical process improvement actions are
shown in Figure 2.
An evaluation method is needed to study which action is most
effective for the
production process. Simulation is used to study the cause-and-
effect relations at this stage.
The simulation results are also used to verify whether the target
is appropriate. If the target
is too stressful, it will go back to negotiate with the higher
decision making level to discuss
the target. Finally, the process improvement solutions can be
generated.
3. Compound workflow model
It is impossible to describe complicated manufacturing
processes from any single
prospective (Toh et al. 1997). Comparing manufacturing
processes with

business processes, the manufacturing process specification
needs to pay much more
attention to constraints such as resource, time and process. A
compound workflow
model (CWM) that includes the process view, resource view and
order view is proposed.
Process level
- Change logic sequence between activities
- Add/delete activities
Activity level
- Change activity processing
- Reducing activity processing time
- Mapping to another resource
Resource level
Management
Level
- Increasing resource capability
- Adding resource’ functions
- Buffering the resource
- Reducing setup time
- Choosing effective scheduling method
Process improvement actions
Figure 2. Typical process improvement actions.

In this section, a brief introduction to the workflow modelling
technique is given to answer
the question why it is chosen. Then the definition of CWM is
discussed in detail.
3.1 Workflow modelling technology
Workflow management is one of the research areas that has
attracted much attention from
researchers, developers and industrial users since the 1990s.
Over the years, there have
been a lot of definitions for workflow and what features a
workflow management system
must provide. For example, the Workflow Management
Coalition (1995) defines workflow
as ‘the computerized facilitation or automation of a business
process, in whole or part’.
Giga Groups call workflow ‘the operation aspects of a business
process, the sequence of
tasks and who perform them, the information flow to support
the tasks, and the tracking
and reporting mechanism that measure and control them’
(Mohan 1997). Fan et al. (2001)
defines workflow as ‘computerized process model which can be
operated by workflow
management system in order to realize business process
integration and automation’.
Two similarities can be found from the above definitions. First,
a workflow model
describes three aspects of the business process, namely:

. ‘What’ is a process? – defining activities that build up the
process;
. ‘How’ is the process organized? – defining logic between
activities;
. ‘Who’ performs the activity? – defining relations between
resources and activities.
Secondly, a workflow model is often associated with an
execution software system.
In other words, it is an executable model that can be read,
operated, and controlled by
a workflow simulation or management system. This unique
character differentiates itself
from other process models.
Workflow management research efforts can be classified into
three categories namely
workflow specification, workflow implementation and workflow
application. A number of
workflow specifications are already available in the literature.
Winograd and Flores (1987)
provided a communication based workflow model by using
speech act theory. It describes
every action of workflow in four phases from a viewpoint of
communication between
customers and performers. An activity-based methodology
which focused on modelling
the work instead of modelling the commitments among humans
is more popular than
a communication based modelling method. The Workflow
Management Coalition
(WfMC) provided a basic process definition meta-model
(Workflow Management

Coalition 1995) which included activity, role and workflow
relevant data. IBM defined
FlowMark (Mohan et al. 1995), which used activities,
input/output containers, connectors
and conditions to describe the business process in build up time
and drove process
instances during run-time. Petri net and its extended form have
also been used for
workflow definition. For example, Ellis and Nutt (1993) defined
information control nets
from Petri net for workflow specification. Van der Aalst (1996)
provided WF-nets, where
transitions presented activities and places described enable
condition of activities.
The workflow management system has often been used in
business process automation
and reengineering (Aversano et al. 2002). It enables process
automation through
integration, coordination, and communication of both human
and automatic tasks of a
business process (Workflow Management Coalition 1996).
Despite applications in
business processes, Lin et al. (2004) studied the job shop
scheduling problem based on
workflow modelling and simulation. The research showed that
workflow models can be
1778 H. Lin et al.
used to describe manufacturing processes and study production
planning and scheduling
problems as well.

As workflow specification helps to clarify process definition,
supports the considera-
tion of production planning and scheduling and simulation
based performance analysis,
workflow modelling and simulation technology has been chosen
in this paper for
manufacturing process analysis.
3.2 Compound workflow model (CWM)
The CWM is proposed especially for the manufacturing process
where resource
constraints description is very important. CWM is composed of
the process view, resource
view and order view (see Figure 3). Each view aims to describe
one perspective of the
process, where multiple relations exist among different views.
(1) Process view
The process view is built using an activity-based method. It is
made up of multi-processes,
each of which defines activities and the process logic needed
for one type of product.
Thus, manufacturing of each type of product has its own pre-
defined production route in
a CWM. A directed acyclic activity-on-node diagram, where
nodes indicate activities and
arcs indicate dependencies, is used for each process. In order to
describe five typical
processes, logic such as serial, and-joint, and-split, or-joint and
or-split logic, and indicate
start/end point of a process instance, six logic nodes are
introduced into the CWM.
For each activity node, there are three kinds of description

namely property definition,
resource mapping and behaviour description.
. Property definition: defines static properties and dynamic
properties of each
activity. The former refers to properties that will not be changed
during operation,
i.e., activity ID and activity’s function. The latter refers to
properties associated
with run-time status such as activities’ begin time, complete
time and real time
priority.
. Resource mapping: allocates resources to activities, including
necessary human
resources and physical resources. It establishes a connection
between the process
Trigger
Activity
Process View
Order
Order View
Enable
Process
Process logic
Individual Resource
Resource View

Behaviour
Description
Resource
Mapping
Properties
Definition
Properties
Resource Pool
Properties Properties
Figure 3. Structure of compound workflow model (CWM).
view and resource view. Cost driver of activities are also
defined via resource
mapping.
. Behaviour description: describes activity’s action by ECA
rules, which are formed
as ‘if Event and Condition then Action’. ‘Event’ refers to
running time events such
as ‘the resource is released’, ‘condition’ refers to activity
enable conditions, and
‘action’ means the activity’s status is transferred from one to
another. It is
behaviour description that enables CWM, has the capability to
describe dynamic

behaviour and makes CWM an executable model. ECA rules
will be explained by
simulation system in the analysis.
(2) Resource view
CWM has an independent resource view so that it can handle
complicated resource
constraints more effectively. Two kinds of resource entities –
individual resources and
resource pools are introduced. The individual resource refers to
a real resource entity that
participates in production. The resource pool is in fact a
classification of individual
resources according to their functions or geographical positions
so that individual
resources in the same resource pool can be substituted for each
other. It makes the model
flexible in dealing with the ‘parallel machine’ problem, which
is very common in
manufacturing.
During the model definition period, static mapping from
individual resources to the
resource pool and resource pool to activities are established
separately. Then during
simulation, individual resources are dynamically allocated to
activities based on static
definition and running time individual resource status. In the
case where there is no
resource pool defined in the resource view, individual resources
will be allocated to
activities directly when building the model.
For an individual resource, properties such as name, function,
capacity, cost, etc. are

considered. For the resource pool, properties such as function
and containing resources
are defined.
(3) Order view
The order view describes properties of orders coming to the
enterprise, which reflects the
market environment that the company is facing. Order instances
are initiated according to
the order view to trigger process instances in simulation.
Properties such as orders’ arrival
time, frequency, type, amount, due date, priority, cost, entry of
the process, etc. are
defined.
(4) Cost specification in the CWM
The cost information is specified in the CWM for two purposes:
first, cost information is
recorded for activity-based costing (ABC). The goal of ABC is
to measure and then price
out all the resources used for activities that support the
production and delivery of
products and services to customers (Kaplan and Atkinson 1998).
CWM provides useful
information for activity based cost (ABC). It clarifies the
activities being performed by the
organization’s resources when establishing the process view.
Then, when the resource is
allocated to the activities, the activity cost drivers and activity
cost driver rate can be
identified. This rate will be used to drive activity costs to
products.
Cost is also an important issue that should be considered in

analysis. Usually the
improvement of the process is often associated with cost
increase. For example,
the decrease of the activity processing time due to the increase
of resource functions or
number of the resources often lead to additional cost of a
resource. Thus, the second
1780 H. Lin et al.
purpose of cost description in the CWM is to record dynamic
changes of activity cost
drivers and their rate in simulation. The process improvement
actions can have impacts on
cost calculating through two ways:
. Spending additional cost on resources, for example, buffering
the resource or
adding special functions to the resource. The increase of the
total resource cost will
lead to the adjustment of the activity cost driver rate.
. Changing the mapping relations between activity and
resources, for example,
mapping the activity to another resource by process
reengineering. In this case,
both cost driver and its rate can be affected.
CWM is going to provide information for activity based cost
analysis (Figure 4). In the

resource view, the resource cost is defined as one of the
resource properties. When the
resource is allocated to the activities, the activity cost driver
and its rate are identified and
recorded as the properties of the activities. During the
simulation, if the adjustment to the
process model is made, it will determine whether the resource
cost and mapping relation
are affected. Then the cost properties of the resource and
activities will be recalculated.
In conclusion, CWM is an adapted workflow model for a
manufacturing process.
Comparing to other process models, it provides the following
features:
. Provides flexible resource definition that support static and
dynamic resource
mapping. It can handle various resource constraints of the
production process.
. Provides process behaviour description by ECA rules. The
model can be used by
the workflow simulation system directly for performance
analysis.
. Provide easy cost specification to calculate activity based cost
and record cost
adjustment in sensibility analysis.
Establish Activity-Based

Process View
Establish Resource View
Allocate Resource to
Activity
Identify activity being
performed
Identify resource in the
manufacturing
Identify activity cost driver
Calculating driver rate
Build order view for
simulation Calculating cost
Building CWM Activity based Cost
Resource cost or
mapping affected?
Process
Improvement
Yes
Figure 4. Cost specification in CWM.
. As mentioned in section 2, workflow modelling and simulation

technology can also
be used for production planning and scheduling problems. That
makes it possible
to support the overall manufacturing process analysis with the
same model.
4. Simulation based sensibility analysis
In order to set the target for the process and find possible
initiatives, the following
questions need to be answered:
. Which factors can have impacts on process performance?
. What kind of impacts the factor can cause to the process?
. How much is the impact?
Sensibility analysis is an effective method to answer the above
three questions. It is
completed by adjusting variables and observing their influence
on the system. However,
there are tens, hundreds or even thousands of variables in
manufacturing. If the variables
are studied one by one, sensibility analysis will become
extremely difficult and
time-consuming. In order to improve the efficiency of the
analysis, a two-phase simulation
method is proposed (Figure 5). The objective of the first round
simulation is to identify
a group of variables that potentially have a major influence on
the system. Then in the
second phase those potential key variables will be studied
further to determine which one
has great impacts on production.

In order to select latent key variables, we first watch the
process’s performance under
different manufacturing environments. A different
manufacturing environment is achieved
by setting appropriate simulation parameters. The following
aspects are considered to
present a different simulation environment:
. Environment factors, including orders’ total number, type,
batch size, due date,
patterns, cost, etc. The process’s performance is usually
associated with the orders
First Round
Simulation
Response to
environments
Influence of different
policies
•
•
•
•
•
Quantitative Cause-
effect-relation
Improvement solutions
Target verification
PSC

Environment
Factors
Performance
Measurement
Statistic
Analysis
Sensibility
Analysis
Group of possible
key variables
Process
Strateg
Process
structure
PSC Environment
Factors
Figure 5. Framework of the two-stage simulation analysis.
1782 H. Lin et al.
coming into the system. For instance, a resource’s utility will be
different when the
order types, pattern, batch size, etc. are changed. A process
improvement solution

should meet requirements under all manufacturing
environments.
. Scheduling policy that is often used in manufacturing.
Different from the
scheduling optimization problem where new algorithms are
studied, the main
purpose of setting a different scheduling policy in the first
round simulation is to
investigate the influence of a company’s most often used
scheduling methods.
It also determines whether the strategy can be achieved by
switching to a different
planning and scheduling method. Since using a different
scheduling policy is more
economic than changing the structure of the process, such a
necessity test is carried
out before structural analysis.
After the data analysis, usually by the ANOVA method, the
crucial performances are
identified. The variables within the process model that are
directly or indirectly associated
with them are going to be studied in the next round of
sensibility analysis. The typical
parameters within the process model are:
. Time related variables, such as processing time of the activity,
set up time, etc.
. Process structure related variables, including the number of
activities, resources,
the topology of the activity network, the association between
activities and
resources.

. Cost related variables, for example, set up cost, holding and
backlog cost, resource
using cost.
. Capability related variables, such as resource capability.
The selected group of process parameters will be treated as
variables in sensibility
analysis. The PSC factors and environment factors are also
considered to give
a comprehensive view of the impacts of these variables under
different scenarios.
5. Example study
A workshop within a small manufacturing enterprise (SME) that
produces crankshafts is
used as a case study. It demonstrates the procedure of applying
the method step by step.
(1) Linking strategy to production process
The company is facing an increasing market and its mid-term
strategy is to increase the
company’s production capability. Currently the workshop’s
output is about 500 items per
month. Assume that the company set a rough target of
increasing manufacturing
capability by 15%. The objective of the analysis is to find the
effective and efficient ways to
increase the throughput.
(2) Build CWM model
The workshop provides several types of crankshafts. Their
manufacturing routings can be

classified into two groups: (1) rough milling – milling – milling
– ultrasound test – drilling –
rough lathing – rough lathing – milling – lathing; (2) rough
milling – milling – ultrasound
test – drilling – rough lathing – rough lathing – milling –
lathing. Four milling machines
(M1, M2, M3, M4), three lathes (L1, L2, L3), one drilling press
(D1) and one ultrasound
test tool (U1) are involved. One operation on one machine is
considered as one activity.
The duration of each activity and the resources used are
described in Table 1.
A two process CWM is built (Figure 6), where each process
describes one
manufacturing route of one product. The resource mapping
between activities and
resources is defined in activity properties.
(3) Simulation based analysis
The following aspects are considered in first round simulation
parameter settings:
. Total order number per month. As the objective is to increase
the process
throughput, the basic rule to set total order number per month is
that it can
cover the situation from current process capability to the target

capability.
Thus, together with the settings of order pattern, the response of
the process to the
manufacturing environment can be watched.
. Order pattern. Three order patterns are considered. Under
pattern A, all the orders
have the same due date for one month. It reflects the stable
market where all the
orders are known in advance and the workshop can arrange the
manufacturing
by a monthly production plan. On the other hand, order patterns
B and C have
10% (B) and 30% (C) urgent orders respectively. The urgent
orders have a due
date of two weeks. Patterns B and C simulate a market with
different degrees of
Table 1. Description of the production route of the case study.
Product 1 Task T1 T2 T3 T4 T5 T6 T7 T8 T9
Machine M1 M2 M3 U1 D1 L1 L2 M4 L3
Duration (hour) 0.3 0.5 0.4 0.5 0.3 0.4 0.5 0.4 0.3
Product 2 Task T11 T12 T13 T14 T15 T16 T17 T18
Machine M1 M3 U1 D1 L1 L2 M4 L3
Duration (hour) 0.3 0.5 0.4 0.4 0.6 0.5 0.3 0.4
T1 T2 T3 T4

T11 T12 T13 T14
P1
P2
T5
Resource View
Activity ID : T13
Activity function : Ultrasound test
Resource: U1
Activity Time : 0.4 hour
Behaviour: If (T13-Previous.Status = completed )
And (Resource.Status = occupied )
Then (T13.Status = initiatived )
If (T13-Previous.Status = completed )
And (Resource.Status = available )
Then (T13.Status = started )
……
Milling
-M1
O-P1
Order
Type:P2
Frequency:
Batch :
Due date:
T6 T7 T8 T9

T15 T16 T17 T18
Process View
O-P2
Milling
-M2
Milling
-M3
Test tool
-U1
Drill Process
-D1
Lathe
-L1
Lathe
-L2
Milling
-M4
Lathe
-L3
Figure 6. CWM model of the case study.
1784 H. Lin et al.

disturbance. According to the historic data, about 65% of the
total order is for
product 1.
. Scheduling policy. The dispatching rules are used in real
production. The three
most used dispatching rules are studied: earliest due date
(EDD), shortest process
time (SPT), and first-in-first-out (FIFO).
The parameters of the simulation are set as in Table 2. Two
performance indicators are
selected:
. Process total throughput which is defined as the number of
units produced per
month.
. Resource utility which is defined as percentage of time a
machine is actually busy.
The result of simulation and ANOVA analysis are given in
Tables 3 and 4 respectively.
The scheduling policy has the highest F-value on the output and
resource utilization
among the three factors. That means the appropriate scheduling
policy can help to
improve the process performance. However, scheduling policy
cannot reduce the demand

on the constraining factors of U1, L1 and L2: in all simulation
runs resources U1, L1 and
L2 have relatively high utilization under all situations. They are
the bottleneck of
production. According to improvement dimensions, activity
level and resource level of
factors are chosen as potential key variables in sensibility
analysis. They are:
. The processing time of activities T4 and T13, T6 and T15, and
T7 and T16, which
are associated with U1, L1 and L2 respectively. Since
technically it is not easy to
dramatically reduce activities’ processing time, two situations
are considered:
reducing processing time 5% and 10%; and reducing the
activities’ processing time
can be achieved through process innovation, using better
material or including
additional functions to the equipment resource (Table 5).
. The number of U1, L1 and L2. One resource will be added.
All three order patterns and three scheduling policies are
considered in sensibility
analysis, it cannot guarantee scheduling policy SPT still has the
best performance, so all
three scheduling policies are considered as well.

Table 2. Description of parameters for first round simulation.
Exp No. Order no.
Order
pattern
Scheduling
policy Exp No. Order no.
Order
pattern
Scheduling
policy
1 525 A EDD 15 575 B SPT
2 550 A EDD 16 525 C SPT
3 575 A EDD 17 550 C SPT
4 525 B EDD 18 575 C SPT
5 550 B EDD 19 525 A FIFO
6 575 B EDD 20 550 A FIFO
7 525 C EDD 21 575 A FIFO
8 550 C EDD 22 525 B FIFO
9 575 C EDD 23 550 B FIFO
10 525 A SPT 24 575 B FIFO
11 550 A SPT 25 525 C FIFO
12 575 A SPT 26 550 C FIFO
13 525 B SPT 27 575 C FIFO
14 550 B SPT
The simulation results of sensibility analysis are shown in Table

6. It can be seen that
SPT still has the best performance among the three scheduling
policies. Although resource
U1 does not have the highest utilization in the first round
simulation, it has an outstanding
performance in the sensibility analysis. The production
capability can be notably
increased by reducing the processing time of T4 and T13 by
10% (production capability
Table 3. Simulation result of first round simulation.
Resource utilization%
Total
throughput M1 M2 M3 U1 D1 L1 L2 M4 L3
1 470 58 62 84 89 64 90 96 70 64
2 469 58 62 84 89 65 91 96 70 65
3 469 58 62 84 89 65 91 96 70 65
4 471 58 63 84 90 65 91 96 71 65
5 472 58 62 84 90 65 91 97 71 65
6 472 59 63 85 91 66 92 98 71 66
7 479 59 63 85 91 66 93 98 71 66
8 478 59 63 85 91 66 92 98 71 66
9 478 57 62 83 88 64 89 95 69 64
10 521 57 61 83 88 64 89 95 69 64
11 524 57 61 83 88 64 89 95 69 64
12 523 57 62 83 88 64 89 95 69 64
13 521 57 61 83 88 64 89 95 69 64
14 524 57 61 83 88 64 89 95 69 64
15 523 57 62 83 88 64 89 95 69 64
16 521 57 61 83 88 64 90 95 69 64
17 524 57 61 83 88 64 90 95 69 64
18 523 58 62 84 89 64 90 96 70 64
19 470 58 62 84 89 65 91 96 70 65

20 469 58 62 84 89 65 91 96 70 65
21 469 58 62 84 89 64 90 96 70 64
22 470 58 62 84 89 65 91 96 70 65
23 469 58 62 84 89 65 91 96 70 65
24 469 58 62 84 89 64 90 96 70 65
25 470 58 62 84 89 65 91 96 70 65
26 469 58 62 84 89 65 91 96 70 65
27 469 58 62 84 89 65 91 96 70 65
Table 4. ANOVA results (F-value) for the interaction effects of
three control parameters on total
throughput and each machine’s utilization.
Resource utilization%
Total
throughput M1 M2 M3 U1 D1 L1 L2 M4 L3
Order no 0.14 0.00 1.52 0.00 0.08 0.96 1.06 0.08 0.16 0.46
Order pattern 4.20
a
0.53 0.87 0.53 1.29 0.96 1.85 0.98 1.08 1.21
Scheduling policy 1535.21
a
14.74
a
16.52
a
14.74
a

13.63
a
10.82
a
14.37
a
11.13
a
17.37
a
11.90
a
a
With a significance less than 5%.
1786 H. Lin et al.
increase 12.65%) or by adding one U1 (production capability
increase 15.95%). It has
much better performance than adding L2 and L1, or changing
the activity time associated
with them. The reason for this perhaps is that U1 not only has
relative high utilization, but
also has been involved in production in an earlier stage than L2
or L1. The target can be

realized by adding one ultrasound test tool.
6. Conclusions
Manufacturing process analysis is an important stage to improve
a manufacturing
enterprise’s performance. It is a higher level of analysis that
focuses on strategic thinking
of the manufacturing and structure redesign of the process with
the consideration of PSC.
Table 5. Reduce activities’ processing time.
Description Through Cost
1 Reduce T4, T13’s processing
time 5%
Process innovation –
time 10%
Use better material T4 and T13’s material cost driver rate
"4%
time 5%
Process innovation –
time 10%
Add L1’s function T6 and T15’s resource cost driver

rate"6%
time 5%
Process innovation T7 and T16’s resource cost driver
rate"2%
time 10%
Adding L2’s function T7 and T16’s resource cost driver
rate"8%
Note: – means not changed.
Table 6. Impacts of select variables on process performance.
EDD SPT FIFO
No. Avg.p IncsR Avg.p IncsR Avg.p IncsR
1 469.29 0.01 562.59 7.52 469.29 0.01
2 469.32 0.01 589.45 12.65 469.32 0.01
3 482.60 2.84 526.65 0.65 482.60 2.84
4 482.73 2.87 525.93 0.51 482.73 2.87
5 469.96 0.15 523.35 0.02 469.96 0.15
6 470.66 0.30 523.43 0.03 470.66 0.30
7 469.26 0.00 606.74 15.95 469.26 0.00
8 482.48 2.82 511.43 2.26 482.37 2.79
9 470.85 0.34 523.26 0.00 471.01 0.37
Note: Avg.p¼average process throughput under all three order
patterns and
IncsR¼Avg.p/current process throughput%.

The authors provided a useful manufacturing process modelling
method and simulation
based analysis framework to support MPA. It has the following
major advantages:
. The compound workflow model (CWM) provides a graphic
process
description for MPA. It introduces rich model elements, for
example, logic
nodes, resource pool, and ECA rules, to provide a flexible
description of
various constraints in the manufacturing process. Also, it can be
used directly
by simulation to study the impacts of scheduling policy and
analyse the
process performance.
. The analysis framework gives a step-to-step guide for MPA
with process modelling
and simulation techniques. Despite the usual steps of setting
objectives, selecting
key performance measures, setting targets and identifying
initiatives, the target
verification is introduced into the framework to build a bridge
between the
decision making level that sets targets and operation level that
have details about
manufacturing.
. The two-step simulation method provides an effective way to
quantitatively
identify cause-and-effect relations needed in MPA. Also, it

provides a general
framework to take the manufacturing environment and impacts
of scheduling
policy into consideration.
Further work will consider the framework to identify, evaluate
and determine the
process improvement solutions when cause-and-effect relations
are quantitatively defined.
More factors, such as quality, may be studied in the framework
with support of modelling
and simulation technology.
References
Altiok, T., 1996. Performance analysis of manufacturing
systems. New York, NY: Springer-Verlag.
Askin, R. and Standridge, A., 1993. Modeling and analysis of
manufacturing systems. New York,
NY: John Wiley & Sons Inc.
Aversano, L, Canfora, G., De Lucia, A. and Gallucci, P., 2002.
Business process reengineering and
workflow automation: A technology transfer experience. The
Journal of Systems and Software,
63, 29–44.
Challis, D., Samson, D. and Lawson, B., 2002. Integrated
manufacturing, employee and business
performance: Australian and New Zealand evidence.
International Journal of Production
Research, 40, 1941–1964.
Cooke, D.L., Rohleder, T.R. and Silver, E.A., 2004. Finding
effective schedules for the economic lot

scheduling problem: A simple mixed integer programming
approach. International Journal of
Production Research, 42, 21–36.
Ellis, C.A. and Nutt, G.J., 1993, Modeling and enactment of
workflow systems.
In: M.A. Marsan, ed. Application and theory of petri nets.
Berlin, Germany: Springer-
Verlag, 1–16.
Fan, Y., Luo, H.B. and Lin, H.P., 2001. Fundamentals of
workflow management technology. Beijing,
China: Tsinghua University Press/Springer-Verlag (In Chinese).
Goldratt, E.M. and Cox, J., 1992. The goal: a process of
ongoing improvement, 2nd ed. New York,
NY: North River Press.
Gupta, Y.P., Evans, G.W. and Gupta, M.C., 1991. Review of
multi-criterion approaches
to FMS scheduling problems. International Journal of
Production Economics, 22,
13–31.
Hammer, M. and Champy, J., 1993. Reengineering the
corporation: a manifesto for business
revolution. New York, NY: Harper Collins.
1788 H. Lin et al.
Kaplan, R. and Norton, D., (Jan–Feb) 1992. The balanced
scorecard-measures that drive
performance. Harvard Business Review, 70, 71–79.

Kaplan, R. and Norton, D., 1996. The balanced scorecard:
translating strategy into action. Boston,
MA: Harvard Business School Press.
Kaplan, R. and Atkinson, A., 1998. Advanced management
accounting, 3rd ed. Englewood Cliffs, NJ:
Prentice Hall Inc.
King, W.R., 1994. Process reengineering: the strategic
dimensions. Information Systems
Management, 11, 71–73.
Lee, D.Y. and DiCesare, F., 1994. Scheduling FMS using Petri
nets and heuristic search. IEEE
Transactions on Robotics and Automation, 10, 123–132.
Lin, H.P., Fan, Y. and Loiacono, T.E., 2004. A practical
scheduling method based on workflow
management technology. International Journal of Advanced
Manufacturing Technology, 24,
919–924.
MacCarthy, B.L. and Liu, J.Y., 1993. Addressing the gap in
scheduling research: a review of
optimization and heuristic methods in production scheduling.
International Journal of
Production Research, 31, 59–79.
Mohan, C., Agrawal, D., Alonso, G., Abbadi, A.E., Gunthor, R.
and Kamath, M., 1995. Exotica:
a project on advanced transaction management and workflow
systems. SIGOIS Bulletin –
Special Issue on Business Process Management Systems:
Concepts, Methods and Technology,
16, 45–50.

Mohan, C., Recent trends in workflow management products,
standards, and research, 1997.
Available online at:
http://www.almaden.ibm.com/cs/exotica/wfnato97.pdf (accessed
28
August 2007).
Potts, G.N. and Kovalyov, M.Y., 2000. Scheduling with
batching: a review. European Journal of
Operational Research, 120, 228–249.
Proth, J.M. and Minis, I., 1995, Planning and scheduling based
on Petri nets.
In: M.C. Zhou, ed. Petri nets in flexible & agile automation.
Norwell, MA: Kluwer
Academic, 109–148.
Shih, H. and Sekiguchi, T., 1991. A timed Petri net and beam
search based on-line FMS scheduling
system with routing flexibility, in Proceedings of 1991 IEEE
international conference on
robotics and automation, Sacramento, CA, April, pp. 2548–
2553.
Sox, C.R., Jackson, P.L. Bowman, A. and Muckstadt, J.A.,
1999. Review of the stochastic lot
scheduling problem. International Journal of Production
Economics, 62, 181–200.
Tan, W. and Khoshnevis, B., 2000. Integration of process
planning and scheduling-a review. Journal
of Intelligent Manufacturing, 11, 51–63.
Toh, K.T., Newman, S.T. and Bell, R., 1997. An Architecture
for Information Specification Based
on an SME Enterprise Model, in Proceedings of the IEE 5th

international conference factory
2000: the technology exploitation process, Institution of
Electrical Engineers, Cambridge, UK,
pp. 456–463.
Tsai, C.C., Chu, C.H. and Barta, T.A., 1997. Modeling and
analysis of a manufac-
turing cell formation problem with fuzzy mixed-integer
programming. IIE Transactions, 29,
533–547.
Van derAalst, W.M.P., 1998. Three good reasons for using a
Petri-net-based workflow management
system, in Proceedings of the international working conference
on information and
process integration in enterprises (IPIC), edited by S. Navathe
and T. Wakayama,
Cambridge, Massachusetts, Nov, 1996, pp. 179–201. [Also in
Information and process
integration in enterprises: rethinking documents. Kluwer
Academic Publishers: Norwell MA,
pp. 161–182.
Viswanadham, N., 1999. Analysis of manufacturing enterprises:
an approach to leveraging value
delivery processes for competitive advantage. Norwell, MA:
Kluwer Academic Publishers.
Winograd, T. and Flores, R., 1987. Understanding computers
and cognition. Reading, MA: Addison-
Wesley.
Workflow Management Coalition, The workflow reference
model, Document number TC00-1003,
1995.

Workflow Management Coalition, Workflow management
coalition terminology and glossary
(Document number WfMC-TC-1011), Technical Report,
Brussels, 1996.
Xiong, H.H., Zhou, M.C. and Caudill, R.J., 1996. A hybrid
heuristic search algorithm for scheduling
flexible manufacturing systems, in Proceedings of 1996 IEEE
international conference on
robotics and automation, Minneapolis, MN, April, pp. 2793–
2797.
Yun, Y.S. and Gen, M., 2002. Advanced scheduling problem
using constraint
programming techniques in SCM environment. Computers and
Industrial Engineering, 43,
213–229.
Zhou, M.C. and Venkatesh, K., 1999. Modeling, simulation, and
control of flexible manufacturing
systems: a petri net approach. River Edge, NJ: World Scientific
Publishing Company.
1790 H. Lin et al.
Economica, 57. 485-93

Pattern of Trade between Underemployed Economies
By RAVEENDRA N . BATRA and HAMID BELADI
Southern Methodist University and University of Dayton
Final version received 6 September 1989. Accepted 13
December 1989.
In this paper, we have analysed the factors determining the
pattern of trade between
underemployed economics. We find, among other things, that a
iow-wage, iand-abundant
country exports the iand-using, labour-intensive product to a
high-wage, capital-abundant
country. We also find conditions under which a high-wage,
capital-rich country may export
the labour-iniensive product (the Leontief Paradox).
INTRODUCTION
The following facts should be kept in mind while formulating a
theory of a
country's pattern of trade.
1. The vast majority of trading countries have suffered from
chronic unemploy-
ment throughout this century. Even many developed economies
such as
the UK and USA experienced low to high unemployment during
the eariy
1920s, throughout the 1930s, at the end of the 1950s, during the
eariy 1960s
and throughout the 1970s and 1980s.
2. Some countries rich in natural resources export land-using

(or land-inten-
sive) products, whereas capital-abundant countries mostly
export capital-
intensive goods, but at times export the labour-intensive
products, some-
thing that has inspired the literature on the Leontief Paradox.
3. In spite of the presence of trade, the capital-rich countries
have higher
wages but lower capital rentals than the labour-surplus and/or
land-rich
countries—a fact conflicting with the factor-price equalization
theorems of
the popular Heckscher-Ohlin model.
4. Labour-rich countries usually export either labour-intensive
or iand-using
commodities.
The purpose of this paper is to formulate a theory of trade that
explains
the 'stylized facts" mentioned above. Neither of the two popular
theories, the
Ricardian model and the Heckscher-Ohlin model, is compatible
with these
facts. Both of these models ignore unemployment and are
couched in terms
of one or two factors, whereas a realistic theory of international
trade must
assume at least three factors—labour, capital and land (or
natural resources)—
which have been incorporated into the empirical literature but
not into a theory
of trade patterns.
In order to explain the pattern of trade, we first construct a two-

sector,
three-factor mode! of an underemployed economy. We find that
the specific-
factor model, which has recently regained its lost popularity, is
ideal to
capturing the facts mentioned above.
L ASSUMPTIONS AND THE MODEL
Let us assume that there are two sectors, X and Y, with X using
capital and
labour and Y using capital, labour and land. (X could represent
all manufactur-
ing where land is relatively insignificant, and Y could represent
agriculture,
486 ECONOMICA [NOVEMBER
mining, forestry and fisheries. I Production functions are
linearly homogeneous;
producers face perfect markets in goods, capital and land, but
not in labour,
where real wage is rigid, causing general unemployment in the
economy.
Capital and labour are fully mobile and employed, and all
factors are in
inelastic supply. In short, we make all the assumptions of the
well-known
two-sector, specific-sector model, except that the real wage is
determined
institutionally."
Let Cij be the ratio of the ith factor and thejth product {i = K, U
^ and

j = x,y) where K is capital, L is labour and V is land. With full
employment
of capita] and land, we have
(1) CK.x + C K , y = ^
and
(2) CvvV'= v;
where K and V are inelastic supplies of capital and land. With
producers
producing in perfect markets, the price of each product equals
the unit cost.
Let Y be the numeraire, so that its price equals one, F be the
relative price
of X, w be the real wage rate, r the real rental of capital and p
the real rental
of land. Then
(3) wC
and
(4) H-C, ,
With linearly homogeneous production functions, each input-
output
coefficient depends only on factor prices. Therefore
(5)
and
With this equation, the production side of our model is
complete. This is a

system of nine variables (X, Y, C:,, C , r and p), nine equations
and four
parameters (w, P, K and V).
II. THE FACTOR-PRICE DEFINITION OF FACTOR
ABUNDANCE
Let us assume that there are two countries, a home country (H)
and a foreign
country (F). Since labour is in excess supply in both countries,
we will first
define relative factor abundance in terms of fully utilized
factors. Suppose the
home country is rich in capital but poor in land relative to the
foreign country.
Then we define H to be capital-abundant and F to be land-
abundant if, in the
absence of trade,
(7)
In other words, in the absence of trade, the capital-rich country
has a higher
rental of land relative to that of capital than does the foreign
country. What
1990] PATTERN OF TRADE BETWEEN UNDEREMPLOYED
ECONOMIES 487
about the wage rate? Since usually the capital-rich countries are
high-wage
economies, we assume that
(8) Wft s wj.

In order to use (7) and (S) in our analysis, we need to analyse
the
relationship between factor and commodity prices. Totally
differentiating (3)
and (4), we obtain
(9)
and
( 1 0) d^yW* + dKyr* + Bvyp* = 0
where the asterisk denotes proportional change and $ are the
factor shares.^
Thus, r* = (dr/r) and so on. From (9) and (10), we obtain
- - « . ^ W (
and
In obtaining (11)-(13), we have made use of the fact that factor
shares add
to one. Thus,
Equation (13) shows the effect of P and w on the factor-price
ratio {pi r), as
( p * - r * ) / P * < 0 . For any given w, a rise in P
unambiguously causes a fall in
ip/r), whereas for any P, a rise in w causes a rise in (p/r),
provided 6i_^> Oi^y.
Since w is constant, a rise in P, the relative price of X, must
cause a rise in r,
otherwise average cost in X will not rise; but since Py = 1 is
constant, p must
fall, because with w given, r and p must move opposite to each

other. Otherwise
average cost in Y will not remain constant. On the other hand,
for any P, a
rise in w must cause a fall in r in X and hence in Y as well.
Since w rises and
r falls, the reward of land depends on the labour intensity of Y
relative to X.
If 0Ky0Lx > ^Lv^Kxy then in the physical sense X is labour-
intensive relative to
Y, and ( P * / R ' * ) > 0 . When the capital-labour ratio in X
and Y is the same,
then a rise in w combined with a fall in r equally affects the unit
cost in Y
which is compatible with the fact that P,, is unchanged at one.
However, if Y
is capital-intensive relative to X, then the relative unit cost in Y
falls, and for
its price to remain constant p must rise. Thus, when
then
( P * / H ' * ) > 0 .
On the other hand, when labour share is higher in Y, so that
0LX < ^LV, a rise
in w hurts Y more than X. Here (p/r) must fall as w rises (see
(13)).
The two forces that affect (p/r) are thus P and w, but not factor
supplies.
Figure 1 displays these relationships. The negatively sloped
curve AB shows

that as P rises ip/r) falls for any w. Then as w rises, the curve
AB shifts up
to CD (where di^ > 0^.), or down to EF (where Bu < ^/.y).
The relationships obtained in Figure 1 can be used to analyse
the effect of
factor endowments on the pattern of trade. Let us assume that
production
functions are the same between the two countries. (This is also
an assumption
of the Heckscher-Ohlin (H-0) theory.) For the time being, let us
also assume
that the wage rates are the same as well; that is, ŵ = w,. Under
these assump-
tions, the same relationship between the factor-price ratio and
the commodity-
price ratio applies lo both countries. Let FH be such a common
curve in
Figure 2. Since the home country is relatively capital-abundant,
its factor-price
ratio (p/r) is placed above the factor-price ratio of the foreign
country
(see (7)). As a result,
(14) P,<P^,
which shows that under autarky the home price of X is lower
relative to that
in the foreign country. This means that, in the presence of non-
intersecting
social indifference curves (which we assume hereafter), the
home country will
export X and the foreign country will expon Y, the land-using
product.' The
following theorem is now immediate.

F
FIGURE I
p/r
Ux
FlQURE 2
ECONOMIES 489
Theorem 1. If the institutionally determined real wages between
the two
countries are close to each other, then the country relatively
well endowed in
land or natural resources exports the land-using product, and the
relatively
capital-rich country exports the other product.
Note that this is a very general theorem, and factor intensities
play no role
in it. This may explain the trade pattern of most oil-exporting
countries, such
as Saudi Arabia, Kuwait, United Arab Emirates, Mexico,
Venezuela, etc.,
which are rich in oil and export this product abroad. Other
countries, such as
Canada, Australia and some Third World countries, export
minerals for the
same reason. Furthermore, it should be noted that the rigid wage
rate may be
'too low' and the country may have to import labour from

abroad. Thus, the
model may also refer to trade between the Arab countries and
the West. Many
of the Arab countries in fact import labour from Asian
countries.
Let us now relax our assumption of similar wage rates between
the trading
partners and assume that the capital-rich country has the higher
real wage
than the land-rich country. If the labour share of the two sectors
is the same,
i.e. if L
̂ x = ^Ly., then w has no effect on (p/r). In this case.
Theorem 1 continues
to hold regardless of the inter-country wage differential.
In most countries the land-using sectors, especially agriculture
and fisheries,
also employ larger number of workers per unit of output than
the other sector.
This is certainly true of developing countries, but may also be
valid with some
developed economies. It is interesting to note that the autarkic
values of (p/r)
depends not only on the levels of endowments of different
factors but also on
the level of the given wage rate. This is clear from equation
(13), where {p/r)
is seen to depend on w and P. The endowment effect in this case
is via the
relative price, P. In other words, even if two countries are
endowed with the
same amounts of the various factors, ip/r) can have different
values in
the two countries. Let us then assume that Bt^y> fl^v Here a
rise in w reduces

{p/r) at any P. If w^ > w^, then the home country's curve
relating {p/r) and
P will lie below FH in Figure 2. Suppose it is given by hh',
whereas in the
foreign country this relationship is still given by FH. Then the
home autarky-
price ratio is given by PJ,. Since it is even smaller than before,
we conclude
that the home country will still import the land-using product
and the foreign
country will export it. All this leads to the following theorem.
Theorem 2. A relatively low-wage, land-abundant country
exports the land-
using product if the latter's labour share is no less than the
labour share of
the other product, which in turn is exported by the relatively
high-wage,
capital-abundant country.
What is the role of the factor intensities in this theorem? Factor
intensities
can be defined in the physical sense or in the value sense. In the
value sense,
the factor intensities can be defined in terms of factor shares.
Thus, Y may be
said to be labour-intensive relative to X if ff^. > L
̂ X- But this
implies that
or
(p/r)

Therefore, if Bi_v > Bi_^, then ê x > B^y, which means that X
is capital-intensive
relative to Y in the value sense. In terms of Theorem 2, a high-
wage, capital-
abundant country exports X when fl^ > L
̂ X or &K,. <^KJC-
Theorem 2 then
implies that a high-wage, capjtal-rich country exports the
capital-intensive
product.
On the other hand, a low-wage country may also be defined as a
labour-
abundant country. Thus, a land-abundant country with lower
real wage may
be said to be a labour-abundant country with lower real wage
may be said to
be a labour-abundant country as well. Theorem 2 then implies
also that a
labour-rich country exports the labour-intensive product, as
Biy> Bj^^. AM this
leads to another theorem.
Theorem 3. A high-wage, capital-abundant country exports the
capital-inten-
sive product, and a low-wage, land- and labour-abundant
country exports the
land-using, labour-intensive product.
This theorem is somewhat reminiscent of the Heckscher-Ohlin
theorem
which highlights the role of factor intensities and inter-country
factor
endowments.
What happens if the land-using product is not labour-intensive

in the value
sense, so that 0^ < ^L. ? Here Theorem 3 may not hold, in
which case the
well-known Leontief Paradox can occur*
If ^L.<^tA, then from (13). {p/r) rises with a rise in w. This
case is
illustrated in Figure 3, where, unlike in Figure 2, hh' lies above
FH., and the
home-autarky price ratio may or may not be below the foreign-
autarky price
ratio. Figure 3 illustrates the case where Ph<P,, so that the high-
wage,
capital-abundant home country will stilt export X but X is now
the labour-
intensive product. Similarly, the low-wage, land-abundant
country exports Y,
which is capital-intensive relative to X. This is the Leontief
Paradox.
Thus, our model, which yields a Heckscher-Ohlin type of
theorem, is also
capable of explaining the Leontief Paradox.
III. FACTOR PRICES UNDER FREE TRADE
Under free trade, the Heckscher-Ohlin theorem gives rise to a
single outcome,
namely that absolute and relative factor rewards are completely
equalized
ECONOMIES 491
between countries. This is one of the most serious flaws of this

theorem,
because we live in a trading world where factor prices are far
from equal.
In our model, however, a variety of outcomes is possible in the
free trade
equilibrium. Let us first consider the least likely outcome.
Suppose, in the
absence of trade, that real wages are the same internationally. In
this case one
common FH curve applies to both countries in Figure 4. Assume
that transport
costs are zero or negligible; then the same product-price ratio
prevails in both
countries in the free trade equilibrium. Let Pg be such a price
ratio. When FH
applies to both countries, then {p/r) is the same in both
countries in the free
trade equilibrium. This is the case where the absolute factor
prices will also
be equalized. Even if wages differ between H and F, the FH
curve applies to
both countries if 0,^ = 6i^,.. In this case (p/r) will be similar
internationally
but absolute factor prices will differ. Since iv̂ > w,, r,, < rj.
This is because r
and w are negatively related. And since {p/r)h = {p/r)f, then
Ph<pf- Thus,
here is a case where relative factor prices of land and capital are
globally
equal but absolute factor prices are not. All this leads to the
following theorem.
Theorem 4. If wages are the same across the countries, then free
trade leads
to an equalization of all factor prices. If the capital-abundant

country is the
high-wage country but labour shares are the same between the
two sectors,
then the relative factor prices of land and capital are equalized
but the capital
rental and the land rent are lower in the capital-abundant
country than in the
land-abundant country.
When SL^ < 6i_y and w>, > w,, much of Theorem 4 continues
to be valid,
although relative factor prices of land and capital in H and F are
no longer
the same. In this case, the home curve representing the
relationship between
ip/r) and P lies below FH, which now represents F. In Figure 4,
the home
curve is now given by hh and in free trade equilibrium, {p/ r)h
is given by OA
which lies below (p/r)^, the factor-price ratio in F. Still, when
w^ > Wf, r,, < Tf.
This is because, from (11), it is clear that
(15) r = r{P,w)
with Tp = {dr/dP) > 0 and r^. = (dr/dw) < 0, for all values of 0^.
and Ot,.. Thus,
with H';, > Wf, rf<r,. Since (p/r),, <ip/r)j in free trade, clearly
Pf<pi, in the
free trade equilibrium.
p/r
B
p/f).

A
F
s
V
^ ^ ^
.
*
FIGURE 4
When ^L, > di_y, the home curve (h'h') now lies above FH in
Figure 4. In
the free trade equilibrium,
(16) {plr),>{pfr)f,
but since ŵ > ny, ^s'^^s-'" view of (16), however, p/may no
longer be lower
than ph under free trade. The following theorem is now
immediate.
Theorem 5. Under free trade, the high-wage, capital-rich
country has a lower
rental of capital than the low-wage country, but the
international relationship
of the land rental depends on factor intensities.

IV. TRADE AND FACTOR PRICES
In the Heckseher-Ohlin model, trade benefits the country's
abundant factor
but hurts its relatively scarce factor. In our model, assuming
that the land-
abundant country exports the land-using product and the
capital-abundant
country exports the other product (which is ensured when 0(.,.
^^(_J, trade
will lower the reward of land and raise the reward of capital in
the home
country, and do the opposite in the foreign country. This is
because in this
case, under autarky, P^ < P,. Therefore, in the presence of
profitable trade,
the relative price of X rises in the home country and falls in the
foreign country.
Consequently r will rise in H but fall in F, and p will fall in H
and rise in F.
TTie following theorem is now immediate.
Theorem 6. When the labour share in the land-using sector is no
lower than
that in the other sector, trade benefits a country's abundant
factor and hurts
the scarce factor.''
It should be noted that Theorem 6 may not be valid if 6^^ >
Bi_y.
V. CONCLUSIONS
Using a two-sector, three-factor model in which land is a
specific factor and

wages are institutionally determined, we have analysed the
factors determining
the pattern of trade between underemployed economies. Our
results are com-
patible with many observed facts. Under the assumptions
usually used in the
H-0 model, we find that a low-wage, land-abundant country
exports the
land-using, labour-intensive product to a high-wage, capital-
abundant country.
On the other hand, if the low-wage country is said to be labour-
abundant,
then a labour-abundant country exports the land-using, labour-
intensive
good and the capital-rich country exports the capital-intensive
good. This is
nothing but a version of the Heckscher-Ohlin theorem derived
in a model
of unemployment.
The paper also derives plausible conditions under which a high-
wage,
capital-rich country may export the labour-intensive product.
This tends to
explain the Leontief Paradox.
Under free trade, if wages are the same across countries, all
factor prices
are equalized. However, if the capital-abundant country is a
high-wage country,
then the reward of capital is higher in the low-wage country
even under free
trade. Thus, unlike the Heckscher-Ohlin model, a variety of
outcomes regard-
ing factor prices in free trade is possible in our model.

1 9 9 0 ] PATTERN OF TRADE BETWEEN
UNDEREMPLOYED ECONOMIES 4 9 3
Finally, we conclude that under realistic conditions trade
benefits the
country's relatively abundant factor and hurts its relatively
scarce factor.*
ACKNOWLEDGMENT
We wish to thank two anonymous referees for helpful comments
to a previous draft
of this paper.
NOTES
1. The earlier trade models of rigid wages have mainly involved
the two-sector, two-factor
Heckscher-Ohlin type of model; see e.g. Das (1981). Yu {1982)
and BhandaH (1986). The
main weakness of this type of model is that it leads to a linear
transformation curve or constant
average costs in the industries. The production side thus yields
limited and unrealistic results.
In an earlier paper (1988), Batra and Beladi have shown that,
even with rigid real wages, the
specilic*factor model used here generates increasing average
cost functions for the two indus-
tries. The input-output coefficient lechnique utilized here
greatly simplifies the analysis. For
another application of this technique, see Lahfri (1983).
2. In obtaining these equations, it must be noted that the
competitive producer, facing given

factor prices, equates the derivative of this unit cost to zero,
thereby minimizing his average
cost. Thus,
^y + n/Q,, + pdCvy = dAC, = 0
where 4 C , is the average cost of Y. Similarly for X.
3. In order to see the importance of this assumption in the
context of the factor-price definition
of factor abundance and the Heckscher-Ohlin theorem, see Batra
{1973, Chapter 3).
4. For the latest analysis of the Leontief Paradox, see Casas and
Choi (1985).
5. See Pattanaik (1974) for an application of this type of
analysis to the effects on savings in a
developing economy.
6. In this paper we have ignored [he well-known physical
definition of international factor
abundance, and have focused only on the factor-price definition.
This is because labour is not
fully employed and does not act as a constraint on the two
outputs. However, a physical type
of definition does suggest itself. We could define country H to
be capital-abundant and country
F to be labour-abundant, if
with
By using a fully fledged production model constructed by the
authors elsewhere (1988), it is
possible to show thai ail our results are valid in terms of the
physical definition as well. In the
interest of brevity, however, we leave this task to the reader.

REFERENCES
B A T R A , R. (1973). Studies in the Pure Theory of
International Trade. New York: St Martin's Press.
——and B E L A I J I , H . (1988). Specific factors,
unemployment and trade theory. WeUwirtschaftliches
Archiv, 124,435-43.
B H A N D A R I , J . S. (1986). Commercial Liberalization in
Less-Developed Countries. New York:
Praeger.
CASA.S, F. R. and C H O I , K, E . (1985). The Leontief
Paradox: continued or resolved. Journal of
Political Economy, 93, 610-15.
D A S , S. P, (1981). Effect of foreign investment in the
presence of unemployment. Journal of
International Economics. I I , 2 4 9 - 5 7 .
LAHtRi, S A J A L (1983). Capacity constraints, alternative
technologies and input-output analysis.
European Economic Review, 27, 147-53.
P A T T A N A I K , P. K. (1974). Trade, distribution and
saving. Journal of International Economics, 4,
77-81.
Yu, E D E N (1982). Unemployment and the theory of customs
union. Economic Journal, 92,399-404.

Copyright of American Economic Review is the property of
American Economic Association and its content
may not be copied or emailed to multiple sites or posted to a
listserv without the copyright holder's express
written permission. However, users may print, download, or
email articles for individual use.
How Does the Global Economic Environment Influence
the Demand for IMF Resources?
SELIM ELEKDAĞ�
The main objective of this paper is to quantify the relationship
between the
global economic environment and Stand-By Arrangements
(SBAs) with the
IMF. The results suggest that oil prices, world interest rates,
and the global
business cycle are the most influential indicators that affect the
number of SBAs
being requested. In addition, the empirical model seems to have
reasonable
accuracy when predicting SBAs. Furthermore, when oil prices,
interest rates,
and the global business cycle are adversely shocked by one

standard deviation,
the conditional probability of an SBA nearly doubles, implying
an increase from
about 6 to 12 SBAs. More critically, the model suggests that
even a steady
deterioration of the global economic climate would imply
increasingly harsher
conditions for developing and emerging market countries, which
may in turn
increase the demand for IMF resources significantly. [JEL F01,
F33, F34, F42]
IMF Staff Papers (2008) 55, 624–653.
doi:10.1057/imfsp.2008.4;
published online 17 June 2008
Considering the favorable global economic environment over
the pastfew years, it is probably not much of a surprise that the
number of
IMF arrangements approved recently is well below historical
averages. But
what—if any—is the link between global economic and
financing conditions
and a country’s potential request for IMF financial assistance?
The main
�
Selim Elekdağ is an economist with the IMF Research
Department. The author is
particularly indebted to Bob Flood, Rex Ghosh, Bikas Joshi,
Ayhan Kose, Alan MacArthur,
Gian Maria Milesi-Ferretti, Alex Pitt, Bjorn Rother, and Juan
Zalduendo for their support
and helpful suggestions.

IMF Staff Papers
Vol. 55, No. 4
& 2008 International Monetary Fund
624
objective of this paper is to rigorously quantify the relationship
between the
global economic environment and the number of Stand-By
Arrangements
(SBAs).
Formal econometric analysis is required to quantify the
relationship
between global economic conditions and the potential demand
for SBAs.
Using panel data techniques, this paper reports results based on
412 SBAs
among 169 members over a period spanning 1970–2004. We
focus on
SBAs because they are the main nonconcessional IMF facility
designed to
provide short-term balance of payments (BOP) assistance to
members.
1
Global activity and liquidity indicators as well as country-
specific factors
were used to identify determinants influencing the number of
SBAs. The
three main global factors affecting the probability of requesting
IMF
financial assistance were found to be oil prices, world interest

rates, and
the global business cycle. The most important country-specific
factors
identified include the member’s real GDP growth, the
depreciation of its
currency vis-à-vis the U.S. dollar, its international reserve
cover, and whether
or not it is an energy exporter. The estimates are robust to
changes in model
specification, as well as choice of global and country-specific
explanatory
variables.
Changes in global economic conditions significantly affect the
probability
of a country’s demand for IMF resources. A scenario in which
the three
global factors are adversely shocked from their respective
averages by one
standard deviation nearly doubles the conditional probability of
an SBA.
Furthermore, when oil prices and interest rates are evaluated at
their
respective historical peaks, and the global business cycle is set
at its deepest
trough in the sample, the conditional probability almost
quadruples to about
14 percent, implying an increase from approximately 6 to 23
SBAs.
The results are intuitive and consistent with economic theory.
Among
other things, a rise in world interest rates may increase a
member’s debt
service costs and limit access to capital markets, higher oil
prices would raise

the import bill (for net oil importers), and a global recession
could decrease
international demand for a member’s exports. More critically,
even if global
economic conditions worsen gradually, the probability of an
approved SBA
increases disproportionately owing to the underlying nonlinear
nature of the
econometric model. Such adverse developments would cause a
deterioration
in a member’s current account balance and could lead to acute
BOP
problems. If a country does not have sufficient access to
international capital
markets, that member may request an IMF arrangement to
mitigate the
consequences of potentially severe macroeconomic adjustment.
The estimated regressions may also be used to predict the
numbers of
SBAs. There are indications that the framework has reasonable
predictive
accuracy. Whereas the actual number of SBAs approved in 2004
was 6, the
model predicts between 5 and 5.7 SBAs in 2004. Furthermore,
out-of-sample
1
See the earlier working paper version of this paper, Elekdağ
(2006).
DEMAND FOR IMF RESOURCES
625

predictions for 2005 ranged between 5.7 and 6.1, whereas the
actual number
of approved SBAs was also 6.
Despite the importance of this topic, research on the empirical
link
between global economic conditions and IMF financing is
scarce. In line with
the survey of Joyce (2004), only Bird and Rowlands (2002) and
Conway
(1994) included global economic factors—which was in both
cases only a
measure of world interest rates. In this context, this paper
builds on the
literature by emphasizing the importance of global economic
conditions and
is also the only study that finds a critical role of oil prices in
the demand for
IMF financial assistance. Even though (in contrast to Bird,
Hussain, and
Joyce, 2004; and Marchesi, 2003) Barro and Lee (2005); Joyce
(1992); and
Knight and Santaella (1997) include time dummies to control
for common
effects of external factors, these frameworks may not be well
suited for
prediction.
2
Further review of the literature also indicates that most of the
studies rely
on short sample periods and therefore miss important events,
including the
financial crises of the late 1990s. In fact, only Barro and Lee

(2005); Bird and
Rowlands (2001); Sturm, Berger, and de Haan (2005); and
Trudel (2005)
include a sample period through at least 2000. Furthermore, as
discussed in
detail below, the country coverage in this paper exceeds that in
other studies,
which could be critical to avoid econometric issues such as
selection bias.
Last, other than this paper, only Barro and Lee (2005) and
Oatley and
Yackee (2000) distinguish among the various types of IMF
facilities.
The results of this paper have relevance for the IMF, for
policymakers
throughout the IMF membership, and for capital market
analysts. The
framework developed in this paper underscores cyclical factors
that are
relevant for future IMF lending capacity. This is especially
important
because unusually harsh economic conditions would likely
imply a bunching
of SBA requests—some of which may be exceptional access
cases. In this
context, this paper is also pertinent for assessing the prospects
for the IMF’s
future income position, which depends on the amount of IMF
credit
outstanding.
I. IMF Arrangements from 1970 to 2004
The IMF is best known as a financial institution that provides
resources to

member countries experiencing temporary BOP problems. The
IMF makes
financial resources available to members in the general
resources account
under a range of policies and facilities, including credit
tranches. More than a
decade after its creation, the IMF developed policies on the use
of its
resources in what came to be known as credit tranches. SBAs
were developed
as the main instrument through which members would access
the credit
2
For example, Barro and Lee (2005) partition their sample into
five 5-year periods,
whereas Knight and Santaella (1997) use an indicator variable
that takes the value of unity
from 1979 to 1991 when using a sample spanning only 1973–91.
Selim Elekdağ
626
tranches, and are available for any BOP need. Access under
SBAs is limited
to 100 percent of quota annually and 300 percent of quota
cumulatively,
although in exceptional circumstances access beyond these
limits has been
granted.
Although the IMF has used a variety of instruments to support

members’
BOP needs, the most utilized facility is the SBA. Figure 1
depicts the number
of SBAs, Extended Fund Facilities (EFFs), first credit tranche
arrangements
(FCTAs), and concessional facilities (the Structural Adjustment
Facility,
Enhanced Structural Adjustment Facility, and Poverty
Reduction and
Growth Facility) against the backdrop of the IMF membership.
3
Table 1
provides the distribution of facilities across selected time
periods. Even
though SBAs historically outnumber other facilities,
concessional IMF
financing is increasing in importance. Although not shown,
during the past
decade exceptional access (especially in response to financial
crises) and
precautionary arrangements have gained prominence, whereas
blended
arrangements have been approved much less frequently.
4
Against this
Figure 1. Facilities and IMF Members
0
5
10

15
20
25
30
35
40
1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990
1992 1994 1996 1998 2000 2002 2004
0
20
40
60
80
100
120
140
160
180
200

SBAs (left scale) EFFs (left scale)
FCTAs (left scale) SAF, ESAF, and PRGFs (left scale)
Members (right scale)
Source: IMF, Policy Development and Review Department
database.
Note: SBA¼Stand-By Arrangement; EFF¼Extended Fund
Facility; FCTA¼First Credit
Tranche Arrangement; SAF¼Structural Adjustment Facility;
ESAF¼Enhanced Structural
Adjustment Facility; PRGF¼Poverty Reduction and Growth
Facility.
3
Appendix I in the working paper version of this paper provides
further details on IMF
policies and facilities, including the various arrangements used
to access IMF credit. See
Elekdağ (2006).
4
The 412 SBAs identified during 1970–2004 do not include
blended arrangements.
627
background, we explore below how global economic and
financial
developments affect the potential demand for IMF resources.

II. Indicators of the Global Economic Environment
Determinants of the global economic environment can broadly
be grouped
into activity and liquidity indicators. Controlling for country-
specific policies
and developments, the main conjecture of this paper is that
world interest
rates, oil prices, and the global business cycle are the most
robust indicators
of the global economic environment that influence the demand
for IMF
financial resources.
5
Interest Rates
Shown in Figure 2 is the U.S. federal funds rate adjusted by
U.S. consumer
price index (CPI) inflation against the backdrop of SBAs during
1970–2004.
Note that with the onset of the Volker disinflation in the early
1980s, both the
real federal funds rate and the number of SBAs reach their
historic peaks.
The parallel movements between SBAs and the interest rate in
the early 1990s
are also noteworthy.
Table 1. IMF Arrangements, 1970–2004
Total 1970–79 1980–89 1990–99 2000–04 1995–2004
Approved 739 166 243 238 92 212

GRA 556 166 195 154 41 114
SBAs 412 93 153 126 40 95
FCTAs 79 62 15 2 0 1
EFFs 65 11 27 26 1 18
SAFs, ESAFs, and PRGFs 183 0 48 84 51 98
Blended arrangements 33 0 25 7 1 4
Source: IMF, Policy Development and Review Department
Stand-By Operations Division
database.
Note: SBA=Stand-By Arrangement; FCTA=first credit tranche
arrangement;
EFF=Extended Fund Facility; SAF=Structural Adjustment
Facility; ESAF=Enhanced
Structural Adjustment Facility; PRGF=Poverty Reduction and
Growth Facility;
GRA=General Resource Account. Approved refers to the total
number of arrangements
approved in the year under consideration. Blended arrangements
are concessional
arrangements (SAF, ESAF, PRGF) combined with an EFF or
SBA to supplement IMF
financial assistance to a member.
5
Appendix Table 3 in the working paper version of this paper
contains a comprehensive
description of the data. See Elekdağ (2006).

Selim Elekdağ
628
Oil Prices
The monthly nominal and real average petroleum spot prices
(APSP) are
displayed in Figure 3.
6
Even though nominal oil prices have reached record
levels, prices adjusted for inflation are still below the peaks of
the late 1970s.
Against the background of SBAs, Figure 4 shows the real APSP
as a
deviation from trend.
7
Note that with the rise in oil prices, there is a trend
increase in SBAs from the mid-1970s until the early 1980s.
With the spike in
oil prices in 1979, the number of approved SBAs more than
doubles,
increasing from 8 in 1978 to 20 in 1979. It is also worth
highlighting how oil
prices and SBAs move in tandem during the 1990s. With the
gradual decline
in the APSP in the mid-1990s, the number of SBAs decreased
from 21 in 1995
to 5 in 1998.

Figure 2. Interest Rates and Stand-By Arrangements
0
5
10
15
20
25
30
35
1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990
1992 1994 1996 1998 2000 2002 2004
-4
-2
0
2
4
6
8
Stand-By Arrangements (left scale)
U.S. real short-term interest rate (right scale)

Sources: IMF, World Economic Outlook and Policy
Development and Review Department
databases; and author’s calculations.
Note: Real U.S. short-term interest rate is calculated by
subtracting U.S. CPI inflation from the
federal funds rate.
6
The APSP is calculated using a simple average of U.K. Brent,
West Texas Intermediate,
and Dubai Fateh spot petroleum prices. The real APSP was
scaled using the U.S. CPI, because
world inflation is contaminated by episodes of hyperinflation.
7
To avoid running spurious regressions, the (log) real APSP is
detrended using a log-
linear trend to ensure stationarity of the real APSP. Deviations
from trend were used rather
than growth rate, for example, to capture the burden of
increased fuel costs more accurately.
Consider Figure 3, which shows that after the 1973 OPEC
shock, when oil prices roughly
tripled, prices did not revert to their original single-digit levels.
The first-differenced series
would not capture this persistence, whereas the linearly
detrended series does.
629

Figure 3. Monthly Average Petroleum Spot Price
(U.S. dollars per barrel)
0
10
20
30
40
50
60
70
80
90
1970M1 1972M5 1974M9 1977M1 1979M5 1981M9 1984M1
1986M5 1988M9 1991M1 1993M5 1995M9 1998M1 2000M5
2002M9
APSP
Real APSP
Sources: IMF, World Economic Outlook database; and author’s
calculations.
Note: APSP¼Average petroleum spot price. Real APSP is
calculated by scaling the APSP by

the U.S. CPI.
Figure 4. Oil Prices and Stand-By Arrangements
0
5
10
15
20
25
30
35
1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990
1992 1994 1996 1998 2000 2002 2004
-150
-100
-50
0
50
100
150

Real Oil Prices (right scale)
Sources: IMF, World Economic Outlook and Policy
Development and Review Department
databases; and author’s calculations.
Note: Real oil prices represented as the deviation from linear
trend, in which average petroleum
spot price (APSP) is calculated by scaling the APSP by U.S.
CPI.
Selim Elekdağ
630
Global Business Cycle
As the main measure of the global business cycle, the deviation
of the
logarithm of real-world GDP from trend is used.
8
Figure 5 displays the
global business cycle with the number of SBAs as the backdrop.
Note that
the two global recessions in the early 1980s and 1990s
correspond to the two
peaks in the number of SBAs approved during 1970–2004.
These figures
provide casual evidence in favor of a link between the global

economic
environment and SBAs, but formal econometric analysis is
required for a
rigorous assessment.
III. Methodology
The objective of this section is to describe the analytical
structure
underpinning the econometric analysis. After the discussion of a
conceptual framework, the section proceeds to discuss the key
determinants that influence the approval of SBAs.
Conceptual Framework
As discussed in Mussa and Savastano (1999), a typical IMF-
supported
program begins with an explicit request from a member. Then
the IMF staff
prepares a blueprint of a program to be used as a basis for
negotiations
between a member’s authorities and the IMF staff. When an
agreement has
been reached, the arrangement has to be cleared by IMF
management and
approved by the IMF Executive Board. This potentially iterative
process
demonstrates how IMF-supported programs depend on joint
decision
making. Using language in line with Knight and Santaella
(1997), a
member’s ‘‘demand’’ for an arrangement, and the IMF’s
‘‘supply’’
(willingness to approve one) are both necessary components of
the process.

9
Most of the literature on IMF arrangements has investigated—
using
binary choice models—the determinants of either participation
in IMF
programs or of program approval in a certain year. Notable
examples of the
former include Joyce (1992); Conway (1994); Vreeland (2004);
and Cerutti
(2007); examples in the latter group of papers include
Przeworski and
Vreeland (2000); Bird and Rowlands (2001); and Barro and Lee
(2005). In
contrast to these studies, Knight and Santaella (1997) explicitly
jointly model
the ‘‘demand’’ and ‘‘supply’’ determinants of IMF program
approval using a
bivariate probit specification. However, they find that a
univariate
specification—which they interpret as a reduced-form demand-
supply
8
The log-linear trend implies an annual real global growth rate of
about 3.4 percent.
9
As emphasized by Cerutti (2007), a similar joint decision
process continues throughout
the life of an arrangement, with the amounts that are finally
drawn while the program is on
track determined in such a manner.

631
model—is superior to the bivariate model.
10
In particular, Knight and
Santaella find that the univariate model predicts the approval of
a financial
arrangement more accurately than the bivariate specification.
11
Therefore,
based on Knight and Santaella, the joint determinants of an SBA
are
modeled using a univariate probit model.
However, the estimation strategy used to uncover the empirical
relationships between global economic conditions and IMF
credit is also
based on two other broad strands of research. The first, based
on
Albuquerque, Loayza, and Servén (2005), uses two sets of
explanatory
variables: global and country-specific. The global variables are
indicators of
the world economic and financial climate, whereas the country-
specific
variables control—among other things—for domestic policies
and
idiosyncratic shocks.
12

The second strand, building on the vast literature
on early warning systems and financial crisis prediction—as
summarized by
Berg, Borensztein, and Pattillo (2004)—regresses a binary
independent
Figure 5. The Global Business Cycle and Stand-By
Arrangements
0
5
10
15
20
25
30
35
1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990
1992 1994 1996 1998 2000 2002 2004
-5
-4
-3
-2

-1
0
1
2
3
4
5
Real world GDP growth (right scale)
Sources: IMF, World Economic Outlook database and author’s
calculations.
Note: Real world GDP measured as the deviation of real world
GDP from a linear trend.
10
One reason may be that many variables that enter the demand
side—for example, BOP
need—are likely to enter the supply side (the IMF’s willingness
to meet that need), thus
complicating the separate identification of the demand and
supply curves.
11
Furthermore, Conway (1994) finds that Tobit and probit
specifications yield similar
results.

Confirmations and ContradictionsThe Leontief Paradox, Reco.docx

Confirmations and ContradictionsThe Leontief Paradox, Reco.docx

Recommended

Recommended

More Related Content

Similar to Confirmations and ContradictionsThe Leontief Paradox, Reco.docx

Similar to Confirmations and ContradictionsThe Leontief Paradox, Reco.docx (15)

More from maxinesmith73660

More from maxinesmith73660 (20)

Recently uploaded

Recently uploaded (20)

Confirmations and ContradictionsThe Leontief Paradox, Reco.docx