AbstractProposals for the restructuring of the Mexican

electricity system have considered the creation of an electricity

market with different structures that range from a pure spot
energy market to a predominantly bilateral market. This paper
presents an static simulation-based approach to study the effect of
bilateral contracting and demand responsiveness on market
power on the Mexican electricity system assuming a market is
implemented. The study considers different supply/demand
growth scenarios and takes into account the potential demand
elasticity according the consumption characteristics in different
areas of the country. The study concludes that due the tight
supply/demand conditions and the weak state of the transmission
system, a full, and unregulated, spot market can observe severe
market power problems. The study shows that, regulation of the
market, demand responsiveness programs and adequate bilateral
contracting levels can avoid the problems foreseen in the
particular market setup, with the static simulation-based
approach to predict such situations.

Index TermsElectricity markets, generation dispatch,
market power, demand elasticity, bilateral contracting.
I. INTRODUCTION
S in many industries that are intensive in capital and
infrastructure, perfect competition in electricity markets is
still a text book condition. The electricity industry has well
known externalities and natural characteristics that makes
perfect competition being far from its realization. Though
worst case scenarios of full market power exercise are also
textbook situations, there is a real need to study market power
in electricity networks. The study of market power and the
development of approaches for its detection and mitigation it
has undergone an increasing research and development. There
are four stages that need be considered when defining a
framework to study market power; these stages are its
definition, its prediction, detection and mitigation. This paper
is related with prediction and mitigation of market power in an
electricity industry that is studying different alternatives for the
opening of the industry to a competition framework under an
specific assumed market structure and design. There is an
increasing literature related to the study of market power in
electricity markets. The definition and measurement of market
power in electricity industries has evolved from HHI indices to
more elaborated benchmark Lerner type measures [1], and
even to dynamical definitions of market power [2]. It is
broadly accepted that market power in electricity systems
needs be defined as the exercise of an ability that a market
participant has in order to increase market prices above its
competitive values. Once a market is functioning, it is always
necessary to study the competitiveness of such market; in [3] a
study of the competitiveness of the California market is
presented, the study uses benchmark comparison of prices. In
[4], a study of the competitiveness of the New England market
is developed using similar but quantity-weighted Lerner
indexes. In [5] a simulation approach is used to study the
California market, the study focuses on the computation of
competitive benchmark prices, the model does not considers
network and inter-temporal interactions of different generation
resources.

There is also an increasing interest on developing analytical
equilibrium models that can model the behavior of agents in an
electricity market and their interactions with the electricity
network. Among these models, there are Nash-Cournot
formulations, Stackelberg models and supply function
equilibrium models; for a review see [6]. This type of models
has potential use on market power prediction or detection.

Most structural and evident market power problems related
to the interactions among agents (supply and demand) and the
electricity network, can be predicted with the use of market
simulation models. In order to predict this type of situations,
worst case scenarios (such as, peak system demand, with zero
hedging for the spot market and totally inelastic demand) can
give good answers. In this work we use such an approach
based on simulation models to study the potential benefits of
demand response and bilateral contracting on mitigating
market power. A market dispatch model where a single market
operator receives bids for demand and supply and bilateral
contracts is assumed. The model is used to study the potential
market power scenarios on a hypothetic market model for the
Mexican electricity system based on such dispatch model. The
study considers different supply/demand growth scenarios and
takes into account the potential demand elasticity according
the consumption characteristics in different areas of the
country. The study concludes that due the tight supply/demand
conditions and the weak state of the transmission system, a
full, and unregulated, spot market can observe severe market
power problems. The study shows that, regulation of the
The Effect of Bilateral Contracting and Demand
Responsiveness on Market Power in the
Mexican Electricity System
Guillermo Gutirrez, Marcelino Madrigal, Francisco de Rosenzweig and J osAguado
A
0-7803-7967-5/03/$17.00 2003 IEEE
Paper accepted for presentation at 2003 IEEE Bologna Power Tech Conference, June 23th-26th, Bologna, Italy

market, demand responsiveness programs and adequate
bilateral contracting levels can avoid the problems foreseen in
the market with the static simulation-based approach.
II. ASSUMED ELECTRICITY MARKET DISPATCH MODEL
Proposals for the restructuring of the Mexican electricity
system have considered the creation of an electricity market
with different structures that range from a pure spot energy
market to a predominantly bilateral market. Even tough the
restructuring process is still on the congressional discussion
phase, and the future structure of a market is not know, we
assume that a central independent system operator will exist,
the operator being in charge of both, market and system
functions. The energy market is assumed a combination of spot
and bilateral transactions.
A. The Mexican Electricity System
The Mexican electricity system consists of more than 500
generation units in around 60 generation stations dispersed in
the Mexican electricity network. A reduced model that
considers the 25 most important interconnections among the
principal generation and load consumption areas is presented
in Figure 1.

Fig. 1. Reduced model of the Mexican electricity system. Most important
generation and consumption areas.

The load in each area is considered as a lumped sum, and
the generation in each of the areas is modeled separately for
each of the types of generation present in the system. This
gives a model which includes 59 generation plants distributed
among the 25 areas considered in the model.
B. The Market Dispatch Model
Standing on the day-before time frame, there is always a
need to have a power dispatch model that will generate the
system dispatch and generate market prices based on the
market model. A simplified model that can be used to
accommodate for both, spot and bilateral transactions, is
considered. The model uses linear bid-functions for suppliers
and consumers, and considers network modeling trough a DC
model. The dual variables of the dispatch model balance
constraints are used to set prices for energy in each of the areas
in the reduced model. The dispatch optimization model is
presented in equation (1).

max
1 1
J I
t t t t
j dj i i
j i
p p
= =

s.t.
( )/ ,
t t t t t t
i bi di i j ij bi
j i
p p p p x i t

+ = +

(1)
t i p
t
p
t
i i
, 0

0 ,
t
dj dj
t
p p j t

t ij p x
ij ij
t
j
t
i
, / ) (


The objective function represents social welfare, that is,
dispatching cheapest bids to supply less elastic demand,
according to their bids. Each of the constrains represents: (i)
area power demand constraints, (ii) generation maximum
capacity, (iii) total requested demand in each area, and (iv)
transmission limitations. The demand constrains (1) consider
the effect of bilateral contracts which are all assumed as physic
compromises that are injected into the transmission network.
III. MARKET POWER PREDICTION SCREENING PROCEDURE

Given the structural setting of the assumed generation and
distribution division in the Mexican electricity system
presented in last section. The principal objective is know to
investigate if structural market power could be exercised at the
level of the transmission network; that is, we are interested on
finding situations where the interactions among supply/demand
and the transmission network can lead to situations where a
particular assumed generation company could still exercise
market power in the system. In order to provide fast and
reliable answers an screening procedure, guided with the
market simulation model, is used to produce worst-case
scenarios that are compared with benchmark measures to
determine is market power could be exercised, the simulation-
based scenarios is a reliable and simple alternative to the still
several equilibrium models that are still undergoing academic
development [6].

The procedure requires first, the computation of a benchmark
prices, and then is compared with observed prices generated
from market power scenarios detected trough the use of a Must
Dispatch Index (MDI) index, whose measure indicates the
level of market power that a supplier has in a particular area
of the system; that is, a local type-of concentration index; the
procedure descried below to predict such market power
situation follows from New England ISO market power
monitoring procedures [7]. The procedure can be summarized
in the following steps:

Step 0. Use the actual characteristics of the system and book
O&M cost to determine the marginal cost of generation for
different demand levels (low, medium and high). The resulting
marginal costs will be used ad benchmark prices.
1

20 21 22
13 14
15
8 9
3
16
23
24
17
25
18
10 11
6
7
12
19
4
1
5
2


Step 1. Compute a must dispatch index (MDI) for each
generation in each area of the system, during peak demand
conditions.
( )
0 100
D j i
j i
i
i
P I P P
MDI MDI
P

- - -
=

(2)

The index is a measure of the power output strictly necessary
of a particular generator in a given area regardless its price.
The area demand is denoted by PD the total possible imports
from transmission lines is indicated by I.

Step 2. For generators whose MDI is considerable high (MDI
> 50) perform a new simulation with the market dispatch
models using increased bid prices (i.e.,
new
=100 X
old
).

Step 3. Determine Lernex indexes to resulting from the
simulation with increased bid prices of generators whose MDI
index is high. Assume in all models fully inelastic demand.

( )
100 0 100
i i
i i
i
p c
L X L
p
-
=


Step 4. Go to next area i=i+1, and return to Step 1.

The situations that are intended to be predicted by this
simple-screening procedure, are those where tight
supply/demand conditions can lead to market power exercise
of a particular supplier in the maker as presented in Figure 2.
During high (and inelastic) demand periods, supply tends to
fall in the short-side, during this situations particular
generators, at particular locations in the network may be able
to unilaterally increase prices and achieve considerable
markups.


Competi tive
Supply
Inelastic demand
Quanti ty
Price
Strategic
Supply
Markup
Competitive price
Market power pri ce

Fig. 2. Market power exercise during tight supply and inelastic/demand
conditions.

Consider, the simple electricity system in Figure 3, where
each generator is 100MW capacity and whose cost in $/MW
correspond to the number inscribed inside each generator.

200 MW 300 MW
3 $/MW
50 MW
100 MW
100 MW
50 MW
100 MW
100 MW
50 MW
[ 50 MW ]
1
2
3
4
5
6
6 $/MW

Fig. 3. Competitive benchmark prices for a simple system. Generator are
100MW with costs inside circles. Transmission limit is 50 MW.

The optimal dispatch leads to the benchmark prices (Step 0)
presented in the same Figure. If we compute MDI indexed
(Step 1) in the right hand side area of the system, we find that
generator 6 MDI index is:

6
300 50 (300 100)
50%
100
MDI
- - -
= =

and therefore has strong possibilities to exercise market power,
in this case due to the transmission constraint (I). A new
simulation (Step 3) with an increased price of 100 $/MW for
this generator is executing, leading to the prices in Figure 4,
with a Lerner index of L=(100-6)/100 X 100 =94 % (Step 4)
which indicates an strong exercise of market power.

200 MW 300 MW
3 $/MW
50 MW
100 MW
100 MW
50 MW
100 MW
100 MW
50 MW
[ 50 MW ]
1
2
3
4
5
6
100 $/MW

Fig. 4. Prices after market power ability was predicted and exercised.

This simple example illustrates the procedure that lead to
the prediction of worst/case scenarios of market power as
depicted in Figure 2. The next step is to compute what would
be the effect of demand response or bilateral containing to
mitigate such situations. Observation in Figure 4, shows that
the absence of competitive situation, generator six needs be
asked to contract at least 50MW of its generation at a
regulated price. Similarly if 50MW of the 300MW demand
can be switched to another consumption period, market power
could be clearly mitigated in this situation. Demand response
and bilateral contracting represent an effective means to
deviate market power situations.

IV. MARKET POWER AND MITIGATION SCENARIOS FOR THE
MEXICAN ELECTRICITY SYSTEM

A. Competitive Benchmark Price Estimation (Step 0)

In order to measure the degree of market power that will be
predicted using the simulation approach, competitive
benchmark prices need be estimated. The competitive

benchmark price is estimated using the dispatch model in (1)
considering three different typical demand scenarios (peak,
medium and low demand conditions) for the system in year
2001. Typical operation costs of each generation plant are
obtained from historical data are used to construct a linear bid
function that is feed into the objective function in (1). System
demand is assumed completely inelastic. With the above
assumption the prices in Figure 5 are obtained for each of the
areas in the system and each demand scenario.

Nodal prices
-
100
200
300
400
500
600
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Node
M
x
$
/
M
W
h
Low Demand MediumDemand High Demand


Fig. 5. Competitive benchmark prices for high, mediumand low demand
conditions.

In this section we describe three of the most important
potential market power scenarios that have been predicted
using as a guide the procedure in Section III; which means,
this market power scenarios are due to tight interaction among
supply and demand in the congested transmission network of
the Mexican electricity system, see Figure 2.
B. Market Power and Mitigation Scenario I

After running computing MDI indexes in different areas of the
systems, node 6, shows that a particular generator, a gas
turbine, has MDI=58% (Step 1).

6
621.6MW
2
621.6MW
5
146.2MW
1
94.5MW
CC
SG
292MW
@500$/MWh
GT
212MW
@273$/MWh
212MW
@273$/MWh
GT
52MW
@500$/MWh
GT
14MW
@500$/MWh
SG
150MW
@527.1$/MWh
100MW
8.3MW
95.5MW
46.3MW
6
621.6MW
2
621.6MW
5
146.2MW
1
94.5MW
CC
SG
292MW
@500$/MWh
GT
212MW
@273$/MWh
212MW
@273$/MWh
GT
52MW
@500$/MWh
GT
14MW
@500$/MWh
SG
150MW
@527.1$/MWh
100MW
8.3MW
95.5MW
46.3MW

Fig. 6. Region of nodes (1), (2), (5) and (6). Demand, supply and power
flows in the competitive situation.
Figure 6, shows the connections of area 6 six neighbouring
nodes. The turbo generator has then, has clearly the ability to
exercise market power which is obtains by performing a new
simulation assuming this generates bids its power (Step 2) at
2000$/MWh. The new, market power prices, obtained with
model (1), clearly indicate that this possibility is real, as shown
in the resulting increased prices in Figure 7.

0
500
1000
1500
2000
2500
Node
$
/
M
W
h

Fig 7. Prices after market power exercise in node 6.

However, bilateral contracting can be a useful tool to mitigate
the market power of this generator or any other. In this
exercise, and observing that the generator that exercised
market power will always be needed to supply 759.05 MW at
peak demand; a bilateral contract of such generator, at a
regulated price, will therefore avoid this market power
scenario. The bilateral contract is considered as describes and
the new prices in the market result as presented in Figure 8.

0
100
200
300
400
500
600
$
/
M
W
h
Node

Fig. 8. Prices after bilateral contracting in considered.
C. Market Power and Mitigation Scenario II

The central region of the system (where nodes 8, 13, 14, 15 &
19 are located, see Figure 1) is another area where there can be
strong market power exercise.


13
6519MW
14
179.98MW
15
4023MW
8
1992.8MW
SG 2217MW
@445$/MWh
GEO
15MW
@202$/MWh
GEO
93MW
@202$/MWh
3043.3MW
50.5MW
19
702.67MW
HID
260MW
@260$/MWh
HID
1295MW
@38$/MWh
HID
613MW
@63$/MWh
CC
218MW
@273$/MWh
VC
218MW
@273$/MWh
DUAL
CC
382MW
@268$/MWh
SG 2250MW
@373$/MWh
294MW
600MW
13
6519MW
14
179.98MW
15
4023MW
8
1992.8MW
SG 2217MW
@445$/MWh
GEO
15MW
@202$/MWh
GEO
93MW
@202$/MWh
3043.3MW
50.5MW
19
702.67MW
HID
260MW
@260$/MWh
HID
1295MW
@38$/MWh
HID
613MW
@63$/MWh
CC
218MW
@273$/MWh
VC
218MW
@273$/MWh
DUAL
CC
382MW
@268$/MWh
SG 2250MW
@373$/MWh
294MW
600MW

Fig. 9. Central region. Demand, supply and power flows after simulations


The following graph shows the result after the steam generator
located in node 8 offers a high price bid and therefore
exercises market power.

0
500
1000
1500
2000
2500
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Node
$
/
M
W
h

Fig. 10. Market power exercise in central region

Similarly, in order to mitigate the market power exercise, it
was considered that the demand in node 13 signs up a bilateral
contract for 2218MW with such generator, which is then not
allowed to bid in the spot market the same quantity.

0
100
200
300
400
500
600
1 2 3 4 5 6 7 8 9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
Node
$
/
M
W
h

Fig. 11. Prices after bilateral contracting in the central region
D. Market Power and Mitigation Scenario III

The region where nodes 7, 10, 11, 12 & 16 are located is also
subject of market power exercise. Even though this region has
several generators and there interconnections between almost
all nodes, the result of a high price bid of one generator can
increase the prices of the region. The following graph shows
the results of market power exercise by coal generator located
in node 12.

0
500
1000
1500
2000
2500
1 2 3 4 5 6 7 8 9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
Node
$
/
M
W
h

Fig. 12. Market power exercise in node 12.

11
3080MW
12
876MW
10
747.6MW
7
559.9MW
SG
98MW
@85.4$/MW
h
SG
375MW
@396$/MWh
GEO
800MW
@375.36$/M
Wh
7
M
W
524MW
16
745.95MW
GT
145MW
@410$/MWh
CAR
2600MW
@161.6$/M
Wh
HID
20MW
@85.4$/MW
h
SG
GT 273MW
@500$/MWh
CC
827MW
@273.3$/M
Wh
1550M
W
GT
36MW
@500$/MWh
GT
CC
200MW
@273.4$/M
Wh
SG
540MW
@487$/MWh
2
5
0
M
W
11
3080MW
12
876MW
10
747.6MW
7
559.9MW
SG
98MW
@85.4$/MW
h
SG
375MW
@396$/MWh
GEO
800MW
@375.36$/M
Wh
7
M
W
524MW
16
745.95MW
GT
145MW
@410$/MWh
CAR
2600MW
@161.6$/M
Wh
HID
20MW
@85.4$/MW
h
SG
GT 273MW
@500$/MWh
CC
827MW
@273.3$/M
Wh
1550M
W
GT
36MW
@500$/MWh
GT
CC
200MW
@273.4$/M
Wh
SG
540MW
@487$/MWh
2
5
0
M
W

Fig. 13. Node 11. Demand, supply and power flows after simulations.

In order to mitigate the market power exercise, demand in
node 12 is asked to contract 600MW of its demand and node
11 signs up a contract for 1081MW. The new resulting prices
are present in Figure 14, as can be seen prices go back to
values close to their competitive values presented in Figure 5.


0
100
200
300
400
500
600
1 2 3 4 5 6 7 8 9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
Node
$
/
M
W
h

Fig. 14. Prices after bilateral contracting in the region where nodes 7,10,11
& 16 are located.

E. Demand Responsiveness

The previous examples showed that benefit of bilateral
contracting in order to mitigate market power. It was
demonstrated that a generator with a relatively low installed
capacity can manipulate the prices of several nodes or can
even propagate to a big part of the system. This preliminary
study is also considering the effect of demand responsiveness
to keep low prices and promote competitiveness in the
electricity system. This simulation considers an initial
scenario where market power is present, see Figure 15.


Pri ce
0
500
1000
1500
2000
2500
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Area
$
/
M
W

Fig. 15. Area prices for a predicted market power scenario.

The simulation model is then feed with the potential load
response assumed without knowing the actual characteristics
of load in areas 20 to 22. Once such demand elasticity is
assumed, trough lower price bids in model (1), the prices in
Figure 18 are obtained. Clearly the prices tend to approach
competitive values. Lerner indexed of both scenarios are
compared in Figure 19, as can be seen, strong local market
power is present in areas 20 to 22 if demand where Lerner
index is close to a 100%. Demand response is able to
considerable reduce such levels of market power.


Pri ce
0
100
200
300
400
500
600
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
$
/
M
W

Fig. 17. Area prices for a predicted market power scenario.


Lerner Index
0
10
20
30
40
50
60
70
80
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
%
Demandresponveness
Marginal Costs
Inelastic demand

Fig. 18. Lerner indexes in each are of the system, after and before demand
response.
V. CONCLUSIONS
The paper presents a study of the effect of demand elasticity
and bilateral contracting on the market power in the Mexican
electricity system. Worst-case scenarios are used to predict
most evident market power situations that arise from tight
supply/demand conditions interacting with the transmission
network. Lerner indexes, with competitive benchmark prices
obtained from a cost-based dispatch, are used to estimate de
degree or market power in the system. The study assumes a
particular structural industry structure that includes fifty nine
independent generation stations. As can be seen with the
results in this paper, even if several companies exists the are
remaining possibilities of individual market power exercise at
very high and inelastic demand condition in the system. The
study shoes that bilateral contracting and increasing demand
response is needs to mitigate market power that could be
present in this worst-case situations.

Market power is a complex issue, whose accurate prediction
and mitigation depends in a series of legal and regulatory
factors that interact with the actual functioning of the market.
The study here presented is intended to shed some light in to
actual the possibilities that can be predicted due to the physical
composition of the system and the interaction of supply and
demand in a congested network during peak demand periods.
Other more complex situations, due to the strategic

interactions of different suppliers may need more complex
modeling assumptions, equilibrium models [6] or agent based
simulation [8] are natural extensions of this study.

If an actual market with the industry structure assumed in the
paper will be implemented, depends on the congressional
decisions, this days there are twelve different proposals for
restructuring the Mexican electricity system. Among the ones
that consider the creation of an electricity system, this study
provides input information that needs be considered once the
actual market design is in place.

VI. REFERENCES
[1] S. Stoft, Power Systems Economics: Designing Markets
for Power: IEEE, 2002.
[2] F. L. Alvarado, "Market Power: A Dynamic Definition,
in Proc. 1998 Bulk Power Systems Dynamics and Control
Conf..
[3] S. Borenstein, J . B. Bushnell and F. Wolak, "Diagnosing
Market Power in Californias Deregulated Wholesale
Electricity Market, POWER Working Paper 064,
University of California Energy Institute, 1999.
[4] J . Bushnell, C. Saravia, An Empirical Evaluation of the
Competitiveness of the New England Electricity Market,
CSEM Working Paper 101, University of California
Energy Institute, May 2002.
[5] S. M. Harvey, W. H. Hogan, "Market Power and Market
Simulations, Working Paper, J ohn F. Kennedy School of
Government, Harvard University, J uly 2002.
[6] B. F. Hobbs, U. Helman, J S. Pang, Equilibrium Market
Power Modeling for Large Scale Power Systems, in
Proc. 2001, IEEE Power Eng. Society Summer Meeting,
Vol. 1., pp 558-563.
[7] D. Gan, D. Boucier, Locational Market Power Screening
and Congestion Management: Experience and
Suggestions. IEEE Transactions on Power Systems, Vol.
17, No. 1, Feb. 2002.
[8] D. W. Bunn, F. S. Oliveira, Agent-based simulation an
application to the new electricity trading arrangements of
England and Wales, IEEE Transactions on Power
Systems, Vol. 5, No. 5, October 2001, pp. 493-503.

VII. BIOGRAPHIES

Guillermo Gutirrez. Received a B.Sc. from the Universidad
Iberoamericana in Mexico City. Currently he is pursuing his
M.Sc. from the Universidad Anhuac, in issues related to
market power and its mitigation. Since 2001, he has been
working for the Energy Regulatory Commission (CRE) in the
Electricity Restructuring Unit. Previously, he worked for the
Ministry of Energy in the Investment Promotion Unit and in
the private sector for Iberdrola and URS Dames & Moore.

Marcelino Madrigal. Received a B.Sc, M.Sc and Ph.D from
Morelia Institute of Technology, UANL in Mexico and the
University of Waterloo, Canada, respectively. He has served
as a consultant and instructor in software development and
electricity markets training for the national electricity company
in Mexico and central America. Has participated as instructor
or panelist in workshops, courses and discussion panels inside
Mexico and several other international IEEE meetings and
conferences on electricity markets and optimization
applications issues. He is associate professor at Morelia
Institute of Technology since 1996, and now he acts as
director of research and regulatory development at the Energy
Regulatory Commission in Mexico. He also teaches at the
economics and industrial engineering departments of ITAM
and U. Anahuac. His main areas of research are optimization
applications for electricity energy systems and markets,
competition and regulation of the electricity industry. Any
idea, statement or conclusion in this paper do not represents
the commissions official positions.

Francisco de Rosenzweig holds a Law degree from the
Universidad Panamericana in Mexico City. He has more than
eight years of experience in the public sector. He worked as
executive assistant to the Director of the Legal Research
Institute at the Universidad Nacional Autnoma de Mxico
(UNAM). He held a position as International Associate at the
firm Cleary, Gottlieb, Steen & Hamilton in New York City,
where he took part in the legal framework for the restructuring,
capitalization and modernization of PEMEX. He also
participated in the public debt issuances of Mexico and
PEMEX. Back in Mexico City, he joined the Ministry of
Energy, where he participated in diverse interdisciplinary
working groups focused on the analysis and promotion of
schemes to finance and capitalize the energy sector. Currently
he is the Director General of the Electricity Restructuring Unit
at the Energy Regulatory Commission. Mr. de Rosenzweig has
published several articles related to the International and
Mexican energy sector. He is currently taking part in the
design of the structural reform proposal for the Mexican
Electricity Industry and in the North American Energy
Working Group.

Jose Aguado. Received his Ingeniero Electrico and PhD
degrees from the University of Malaga (Spain) in 1997 and
2001, respectively. He was a visiting researcher at the
University of Waterloo. He is currently Assistant Professor in
the Department of Electrical Engineering at the University of
Malaga. His research interests include operation, planning and
deregulation of electric energy systems and numerical
optimization techniques.