Energy Transition Policy of the Mexican Electricity Sector: A Representation in Terms of the Theory of Change

Abstract

Energy transition in the electricity sector is central to limiting global warming to 1.5 °C. Countries around the world rely on official planning documents, such as strategies, special programs, and action plans, to guide these efforts. However, there are few systematic approaches and tools that can help policy makers determine comprehensiveness, coherence, consistency, and coordination when designing these planning documents. This represents a risk to meet energy transition and climate change goals if the relationships of policies and actions included in these documents are not addressed properly. To this end, this paper proposes an approach to model hierarchical relationships using the theory of change, while causal relationships are allocated using policy makers’ experience (yet can also be exogenously allocated using other models). As a case study, this work investigates Mexico’s energy transition policy of the electricity sector over the 2013–2018 period. The results show that Mexico’s official planning documents lack a good design since most policies and actions are not articulated in the right sequence and privilege clean centralized and distributed generation and energy efficiency, while others related to the modernization of electrical grids and climate and environmental policies are barely included.

1. Introduction

There is consensus in the scientific field that the maximum threshold to avoid unacceptable risks of climate change (CC) would be an increase in the planet’s average temperature between 1.5 °C and 2 °C [1]. To fulfill these efforts, the International Renewable Energy Agency (IRENA) has pointed out the need for tripling renewable power to reach more than 11,000 gigawatts (GW) of global renewable power generation capacity and doubling the global average annual rate of energy efficiency improvements, from around 2% to more than 4% per year between 2023 and 2030 [2]. Moreover, the United Nation’s Environment Program (UNEP) has identified that annual global greenhouse gas (GHG) emissions need to be reduced by 45% compared to current policy-based projections and must continue to decrease rapidly after 2030. Similarly, countries must review and increase their emissions reduction targets for 2030 [3]. Therefore, it is important to ensure that this shift from fossil fuels to renewable energy, usually referred to as energy transition (TE, for its Spanish acronym), goes hand in hand with ambitious actions for climate change mitigation. In this regard, there are countries viewed as world leaders, such as Germany [4], Denmark [5,6,7], the United States of America (California) [8], Costa Rica [9], and Uruguay [10], which have ambitious goals and targets supported by official planning documents (OPDs), such as strategies, special programs, and action plans.

The theory of change has been widely used worldwide in various problems where a system approach is necessary (see, for example, Ref. [11]). In contrast, there are few sectoral planning studies between practitioners of the ToC intended to generate coherent and robust energy policies, especially towards low-carbon energy systems.

For example, Ref. [12] developed a ToC about the role of green financial sector initiatives in the low-carbon transition from coal and carbon-intensive regions in the context of the European net zero climate objective. Ref. [13] considers that the ToC applied to the evolution of energy legislation allows for an analysis on the drivers of change and their linkages from the international to the local level. Ref. [14] analyzed the articulation of Royal World Laboratories and the ToC to improve their performance and reduce energy consumption in the residential sector of several European countries by emphasizing the role of the context and the mechanism of change. However, it was not possible to find any application of the ToC to address TE in the electricity sector, even though this sector is one of the most important emitters of GHG in the world.

The case study proposed in this paper is different than these approaches in that it uses the ToC to investigate hierarchical relationships in OPDs, or, in other words, to what extent policies and actions have been covered by these planning documents in five areas that are assumed to be central for energy transition while interacting hierarchically in terms of the subordination of these planning documents. Therefore, this approach offers policy makers a simplified and flexible way to systematically model hierarchical and causal relationships based on their experience (yet more complex interactions can be handled exogenously with other models) by using commercial software, such as Microsoft Access V.16.0.4266.1001 and Power BI V. 2.131.1203.0.

The originality and purpose of this work are then twofold: (1) to provide policy makers with a model that can be replicated to evaluate hierarchical relationships of policies and actions included in any OPD by using the ToC; (2) to analyze and make recommendations to improve OPDs by using Mexico’s 2013–2018 energy transition policy in the electricity sector (PNTE-SEM, for its Spanish acronym) as a case study. This approach allows policy makers to systematically represent hierarchical and causal relationships based on their own experience or other models with more complex interactions, avoiding fragmented representations that make it difficult to see these relationships when OPDs are analyzed separately.

2. Theory of Change

Studies on change were first introduced at the beginning of the last century and have evolved in the understanding of the type of change, the process of change, and the elements of change since then. The contemporary ToC was developed in the 1990s by the Aspen Institute Roundtable on Community Change [15]. There are several interpretations of the ToC, but all of them try to represent an articulation of policies and corresponding actions to achieve desired outcomes [16]. All in all, it must be ensured that the ToC works in a consistent, coherent, and interrelated way [17].

The term “theory of change” refers to the construction of a model that specifies (usually visually) the underlying logic, assumptions, influences, causal linkages, and expected outcomes of a development program or project. Through the collection and analysis of performance data, this model can be tested against the actual process experienced and results attained by the intervention [18]. There is no set or standardized ToC method, but there are common elements in the precise activities, sequence, and level of detail that are recommended, as described in the following steps (see Figure 1) [19]:

• Identify the goal (sometimes called a “long-term outcome” or “vision”). It should be a clear and specific statement of the proposed outcomes (goals and objectives), and it does need to be clear enough to serve for future planning.
• Develop a pathway of change. It is the centerpiece of the ToC and entails identifying and sorting all the preconditions related to the ultimate outcome into a pathway of change that moves linearly and chronologically from short- to long-term goals. It should be designed via backwards mapping, or, in other words, it should start from the goal, and then systematically work backwards in time from it. Then, it is required to identify specific steps (intermediate outcomes) along the way and respond to the question of “What are the preconditions for the outcomes at this step?” The answers to this question become the outcomes that come before. It is worth emphasizing that the intermediate outcomes that link the early (specific preconditions) and long-term steps should be as strong and robust as possible.
• Define interventions and assumptions. This refers to the way steps are plotted along a pathway of change, and they are framed by assumptions about what the necessary and sufficient conditions to achieve success are. These assumptions have a diverse set of roots, including personal values, professional experiences, evidence and research, and analysis of the overall context or environment.

However, Ref. [20] proposes that the ToC must move forward to become a useful transformation tool for society by articulating the results, demonstrating their feasibility, identifying best practices, and presenting a detailed logic model. Finally, Ref. [21] considered that the ToC can be an effective instrument to support adaptive planning, implementation, learning, and project evaluation carried out by companies, organizations, and governments if integrated into a process of critical reflection and learning.

3. Methodology Used for the Theory of Change of Mexico’s National Energy Transition Policy in the Electricity Sector

3.1. Core Assumptions in the Frame of the Theory of Change

As for the assumptions on how and why a desired change is expected to happen in a particular context, the ToC of the PNTE-SEM is built upon the evaluation of policies and corresponding actions included in the most important OPD within the 2013–2018 period. In this context, it is worth mentioning that after the 2013 constitutional energy reform [22], the organization of the SEM changed, on the one hand, by incorporating a wholesale electricity market and, on the other, by horizontally and vertically disintegrating the Federal Electricity Commission into state-owned productive companies.

Furthermore, this article assumes that not only improving policy design in OPDs under drivers that promote TE worldwide, but also setting goals for clean energy (EL, for its Spanish acronym) and energy efficiency (EE) are essential for any PNTE, especially in the electricity sector.

3.1.1. Drivers of Energy Transition in Electricity Sectors Worldwide

Energy transition can be understood as a change in the primary form of energy consumed [23] resulting from changes in the specific patterns of quantities and qualities of both the supply of and demand for energy. These changes lead to objectives such as reducing consumption or making more efficient use of energy and the increasing use of cleaner energy, such as renewable energy, considering that this transition is more decentralized and sustainable and contributes to reducing inequalities [24,25].

International experience suggests that there are five areas or drivers of energy policy that contribute to TE in the electricity sector:

• The expansion of clean centralized generation (GCL, for its Spanish acronym), i.e., generation at large-scale power plants, which are interconnected to transmission grids and where EL is used in a sustainable manner, especially RE.
• The expansion of clean distributed generation (GDL, for its Spanish acronym) through small power facilities (mainly sustainable RE technologies), which are close to end-users with or without interconnection to the distribution grids.
• The expansion, strengthening, and modernization of the transmission and distribution grids (RTYD, for its Spanish acronym) to massively accommodate grids connected GCL and GDL, especially variable RE, including intelligent and storage systems for optimal management.
• The large-scale promotion of EE, consisting of energy saving and the efficient use of energy in both supply and demand, as well as in RTYD.
• The large-scale reduction in GHG and the negative impacts on health and the environment (hereinafter called health, environment, and climate change—SACC, for its Spanish acronym) by significantly reducing the use of fossil fuels and increasing on a large-scale the sustainable use of EL, especially RE. This contributes to fulfilling or increasing the ambition of GHG reduction goals in the electricity sector.

3.1.2. Official Planning Documents and Their Role in the Energy Transition of the Mexican Electricity Sector

According to Article 21 of the LTE [26], the following OPDs are assumed as the guiding elements for TE under SEM’s structure (see Figure 2): (1) Transition Strategy to Promote the Use of Cleaner Technologies and Fuels (hereinafter referred as the Strategy) [27,28], (2) Special Energy Transition Program (PETE, for its Spanish acronym) [29], (3) National Program for Sustainable Energy Use (PRONASE, for its Spanish acronym) [30], and (4) Energy Sector Program (PROSENER, for its Spanish acronym) [31].

Thus, the role and relationship of policies and OPDs are defined in LTE [26] as follows:

• Article 27 establishes that “the Strategy constitutes the mid- to long-term guiding instrument for national policy on clean and sustainable use of energy, improvements in energy productivity, and, where appropriate, the economic reduction of polluting emissions from the electricity industry...”. This article also establishes that the main objectives of the Strategy are as follows: “I. Set the goals for clean energy and energy efficiency, as well as a roadmap for its implementation; II. Promote the reduction in polluting emissions originated by the electricity industry; and III. Reduce, under criteria of economic viability, the country’s dependence on fossil fuels as a primary source of energy”.
• Article 4 requires that the Strategy “shall set goals so that electricity consumption is met through a portfolio that includes energy efficiency and a growing share of clean energy generation”. Moreover, it requires that Mexico’s Energy Secretariat (SENER, for its Spanish acronym) “will promote that electricity generation from clean energy sources is consistent with related provisions included in the General Law on Climate Change for the electricity industry”.
• Article 5 states that “... the Strategy will establish policies and measures to promote renewable energy use, as well as the substitution of fossil fuels in final energy consumption”.
• Article 33 states that PETE “will establish the activities and projects in line with the Strategy during a government term”.
• Article 35 states that “PRONASE shall establish the actions, projects, and activities in line with the Strategy so as to achieve energy efficiency goals …”.
• Article 25 requires that “... the corresponding sectoral programs shall be in line with policies, programs, actions, and projects included in the Strategy and in the other planning instruments provided for in this Law”.

Furthermore, according to the Planning Law [32] and its latest Amendment to Article 16, section III [33], SENER is responsible for the elaboration of the Energy Sector Program (PROSENER) as the guiding document over a government term (6 years) of the country’s energy policy by setting objectives, priorities, and policies within the framework of the National Development Plan (PND, for its Spanish acronym). Therefore, PROSENER must be in line with the Strategy, PETE, and PRONASE.

To best illustrate these assumptions, Figure 2 shows the relationship of OPDs in accordance with the constitutional and regulatory mandate, while Figure 3 shows the medium- to long-term goals set by the Strategy, as well as the short-term goals set by PROSENER, PETE, and PRONASE for EL and EE.

3.1.3. Selected Analysis Period for Mexico’s National Energy Transition Policy in the Electricity Sector

Since this analysis requires that all the OPDs described in the previous section are implemented, this paper selected a timeframe between 2013 and 2018 to analyze the status of PNTE-SEM. This is due to the fact that over the 2019–2024 period, apart from the Transition Strategy to Promote the Use of Cleaner Technologies and Fuels, which was updated in 2020 [34] and 2024 [35], the other OPDs have faced different setbacks. For instance, the 2020–2024 Energy Sector Program was published in 2020 [36] but was later halted by a definite suspension on all TE-related provisions ordered by a federal court [37]. On the other hand, the 2020–2024 National Program for Sustainable Energy Use was published after 3 years of delay in 2023 [38], whereas the 2017–2018 Special Energy Transition Program had not been updated by the first half of 2024. For this reason, the OPDs published in the 2013–2018 period are best suited for the analysis presented in this work.

3.1.4. Drivers or Areas of Mexico’s National Energy Transition Policy in the Electricity Sector

According to Article 2 of the LTE [26], the following sections are assumed to be consistent with the areas or drivers that promote TE in electricity sectors worldwide: “I. Anticipate the gradual increase of clean energy participation in the electricity industry in order to meet with clean energy generation and emissions reduction goals; II. Facilitate compliance with clean energy and energy efficiency goals established in this Law based on economic feasibility; III. Incorporate externalities in the evaluation of costs associated with the operation and expansion of the electricity industry, including those on health and the environment...; VII. Support the objective and goals established in the General Law on Climate Change to reduce gas emissions and greenhouse effect compounds and to increase electricity generation from clean energy sources; and IX. Promote the energy use of renewable resources and waste...”. Figure 4 shows the linkage between the areas or drivers of TE and the goals set by the OPDs. As can be seen, SACC, EE, and RTYD have an impact on goals for EL and EE, while GCL and GDL promote the utilization of EL, especially RE.

3.2. Representation of the Theory of Change from Official Planning Documents

Based on the frame of the ToC and the core assumptions presented in the previous sections, Figure 5 shows a representation of PNTE-SEM so that an analysis on the design of the OPDs can be evaluated. This structure is made up of the following elements:

• Ultimate outcome: It is at top of the ToC’s structure and represents the ultimate goal, i.e., the PNTE-SEM. As previously mentioned, TE relies on two important elements, a growing participation of EL in the electricity sector and an increasing adoption of EE to comply with established goals, as provided in sections I and II of Article 2 of the LTE [26]. Likewise, sections III, VII, and IX of Article 2 of the LTE require the incorporation of externalities in the cost of electricity generation, besides other aspects that either contribute or are related to PNTE-SEM.
• Conditions: These are at the second highest level of the ToC’s structure and constitute the necessary conditions to reach the ultimate outcome. They refer to mid- to long-term goals on EL and EE set by the Strategy, as well as the short-term goals on EL set by PROSENER and PETE, and EE goals set by PRONASE (see Figure 3);
• Preconditions: These are policies and corresponding lines of action (hereinafter referred to as LAs) in the Strategy and PROSENER, classified by areas or drivers of TE, which must be implemented to bring about either the outcomes or the next precondition in the pathways of change. Two types of preconditions are assumed: (1) initial preconditions and (2) final preconditions. The first ones refer to LAs foreseen in PROSENER, since it is a planning document intended for a government term (6 years). Final preconditions refer to LAs of the Strategy, as they have a sequence closer to the fulfillment of the conditions so that the initial preconditions (short-term) are met first and then the final preconditions (mid- to long-term).
• Specific preconditions: These are at the base of the ToC’s structure, i.e., LA contained in PETE and PRONASE, classified by areas or drivers of TE. These elements are assumed as the interventions for the fulfillment of the preconditions that will in turn contribute to meet short-, medium-, and long-term goals (conditions). Due to the specific nature granted by law to PETE and PRONASE, this article assumes that there is no relationship between their LAs.

3.3. Modeling and Systematization of the Theory of Change

As a first approach (See Figure 6), this analysis focuses on a quantitative assessment of LAs and their relationships to OPDs, or, in other words, the existing relationships between specific preconditions of PETE and PRONASE and preconditions of PROSENER and the Strategy. It also focuses on the relationship between the initial preconditions of PROSENER and the final preconditions of the Strategy. Thus, the following three types of relationships that are relevant for this work are analyzed:

• Policies with strong intensity connections, i.e., LAs related to each other in three OPDs (two connections), namely, the Strategy, PROSENER, and either PETE or PRONASE;
• Policies with moderate intensity connections, i.e., LAs related to each other in two OPDs (one connection), namely, the Strategy and/or PROSENER and either PETE or PRONASE;
• Policies with weak intensity connections, i.e., LAs that can be found in one document only, namely, the Strategy, PROSENER, PETE, or PRONASE; hence, OPDs cannot be related through any of these LAs.

On the other hand, as a second approach, this analysis focuses on a quality assessment of LAs and their relationships to OPDs. In other words, it investigates if a sequence along the connections between OPDs is adequate in terms of reaching the next precondition, as established by the ToC of PNTE-SEM, and the extent to which each sequence has been completed. Thus, the analysis validates if the specific preconditions of PETE or PRONASE contributed to the achievement of either the initial precondition of PROSENER or directly to the final precondition of the Strategy. Similarly, it also validates if the initial precondition of PROSENER and the final precondition of the Strategy are consistent. As a result, the LAs are analyzed by assigning a typology, as described in

Table 1. Typology assumed by the theory of change of Mexico’s energy transition policy in the electricity sector.

ID	Typology	Description
T1	Adequate and fully completed sequence	Sequence is assumed to be correct and fully connected.
T2	Adequate but incomplete sequence	Sequence is assumed to be correct, as in T1, but partially connected.
T3	Inadequate and incomplete sequence	Sequence is assumed to be incorrect and partially connected.

Source: authors’ own elaboration.

and shown in Figure 6.

It is important to note that while this approach models hierarchical relationships in the frame of the ToC, more complex interactions in causal relationships must be analyzed exogenously and then incorporated into the model.

3.3.1. Modeling Hierarchical Relationships

This work uses a branch of mathematical logic known as “set theory” to model the relationship of LAs with the specific preconditions of PETE and PRONASE and the preconditions of the Strategy and PROSENER. This approach has several advantages to ease systematization of this case study.

Concepts and Notations in Set Theory

Let A and B be two sets, whose cartesian product A × B is given by Equation (1):

A = {a₁, a₂, a₃…a_m}; B = {b₁, b₂, b₃…b_n}

(1)

where

A × B = {(a_m, b_n)| a ϵ A ∧ b ϵ B}

(2)

Thus, a binary relation R is the subset of the elements that satisfy a condition such that

R = {(a_m, b_n) ϵ A × B}

(3)

Likewise, let R₁ be a relation of A in B and R₂ be a relation of B in C,

where

C = {c₁, c₂, c₃…c_n}

(4)

Therefore, the composition of R₁ to R₂ (R₁oR₂) is called a new relation of A in C, such that

R₁oR₂ = {(a_m, c_n) ϵ A × C|∃ b ϵ B, (a_m, b_n) ϵ R₁ ∧ (b_m, c_n) ϵ R₂}

(5)

Additionally, the union (U), the intersection (∩), the set difference (–), and the complement (S^c) are operations with binary relations, where O and P are subsets of A × B and can be represented in Venn diagrams, as shown in Figure 7.

Modeling Relationships Between Official Planning Documents

Let A, B, and C be the sets that represent the OPDs, called Strategy (EST), PROSENER (PROS), and Special Programs—PETE and PRONASE (PROGS)—where a, b, and c are elements of A, B, and C, respectively, and represent LAs such that

EST = A = {a₁, a₂, a₃…a_m}; PROS = B = {b₁, b₂, b₃…b_{m or n}};
PROGS = C = {c₁, c₂, c₃…c_n}

(6)

Then, a binary relation R between two OPDs (2R), represented by the subsets O, P, Q, and their LAs (moderate intensity connections), is given for each OPD by

R_{EST_PROS} = {(a_m, b_n)} = O; R_{PROS_PROGS} = {(b_m, c_n)} = P; and R_{EST_PROGS} =
{(a_m, c_n)} = Q

(7)

and can be expressed as illustrated in Figure 8 and Equations (8)–(10).

LA_EST2R = O ∩ P^c ∩ Q^c;

(8)

LA_PROS2R = P ∩ O^c ∩ Q^c;

(9)

LA_PROGS2R = Q ∩ O^c ∩ P^c

(10)

Similarly, LAs with relationships in three OPDs (3R) (Equations (11)–(14)) and in at least two OPDs (∃1R) (Equations (15)–(17)), as well as LAs with no relationship (noR) (Equations (18)–(20)) can be expressed for each OPD in Venn diagrams, as shown in Figure 9.

LA_EST3R = O ∩ (Q ∪ P) ∩ (P^c ∪ Q^c)

(11)

LA_PROS3R = P ∩ (Q ∪ O) ∩ (O^c ∪ Q^c)

(12)

LA_PROGS3R = Q ∩ (P ∪ O) ∩ (O^c ∪ P^c)

(13)

LA_EST3R = LA_PROS3R = LA_PROGS3R = O ∩ P ∩ Q

(14)

LA_EST∃1R = O ∪ (P ∪ Q)

(15)

LA_PROS∃1R = P ∪ (O ∪ Q)

(16)

LA_PROGS∃1R = Q ∪ (O ∪ P)

(17)

LA_ESTnoR = O^c ∪ (P ∪ Q)^c

(18)

LA_PROSnoR = P^c ∪ (O ∪ Q)^c

(19)

LA_PROGSnoR = Q^c ∪ (O ∪ P)^c

(20)

3.3.2. Modeling Sequences Between Official Planning Documents

Figure 10 illustrates a schematic representation of LAs with strong intensity connections. In the frame of the assigned typology along connections between OPDs (See Table 1 and Figure 6), a T1 sequence corresponds to the LAs, where the specific preconditions of either PETE or PRONASE contribute to the achievement of the initial precondition of PROSENER and then to the final precondition of the Strategy, or, in other words, a sequence from the primary or main OPD to a secondary or subordinate OPD (See

Table 2. Examples of policies with strong intensity connections and T1, T2, and T3 sequences.

OPD	Typology				LA
Strategy			↑	T1	Develop universal energy access programs aligned with the United Nations program “Sustainable Energy for All” (SE4ALL).
PROSENER	T1	↑		T1	Advance in the electrification of rural towns and popular neighborhoods.
PETE	T1				Develop renewable rural electrification projects with the participation of communities.
Strategy			↓	T3	Establish large-scale programs to develop national capacities for energy management system implementation and certification in the industrial sector.
PROSENER	T2	↑		T3	Coordinate actions and programs that promote the efficient use of energy, implementing both sustainable production and consumption initiatives.
PRONASE	T2				Develop pilot projects that promote the design of programs to enhance sustainable energy actions in various sectors.
Strategy			↓	T3	Strengthen financing programs to acquire energy efficiency or renewable energy technology at new or existing facilities in the building sector.
PROSENER	T1	↑		T3	Coordinate actions and programs that promote the efficient use of energy, implementing both sustainable production and consumption initiatives.
PETE	T1				Promote financing schemes for clean sources with the participation of development and private banks.

Source: authors’ own elaboration.

). Therefore, these are assumed as LAs with the ideal sequence in the frame of the ToC.

On the other hand, as also shown in Table 2, LAs in a T2 sequence contribute partially to the next hierarchy in the frame of the ToC. In other words, even though the sequence is adequate, as in T1, the connection between OPDs is assumed to be incomplete (this work assumes that LAs in either PETE or PRONASE must be specific actions, while those of PROSENER and the Strategy must be general actions, but not the opposite).

As for a T3 sequence, LAs are assumed to have inadequate sequences when either those included in PETE/PRONASE are not subordinated to the preconditions of PROSENER or when LAs included in PROSENER are not subordinate to the Strategy (this work assumes that PROSENER’s short-term policy must contribute to the Strategy’s medium- to long-term policy, but not the opposite). Additionally, LAs are assumed to be incomplete when expressed, for instance, in terms of a sectoral scope in the Strategy rather than in terms of a broader scope. Similarly, LAs are also assumed to be incomplete when expressed as specific actions in PROSENER rather than as general actions (See Table 2).

Finally, Figure 11 illustrates a schematic representation of LAs with moderate intensity connections, while

Table 3. Examples of policies with moderate intensity connections and T1 and T3 sequences.

OPD	Typology				LA
Strategy			↓	T3	Develop technical regulations and standards for safety, ecological balance, and environmental protection for the comprehensive administration of sustainable geothermal systems.
PROSENER				T3	Promote the conditions for the sustainable use of the country’s water and geothermal resources.
PETE/PRONASE
Strategy			↑	T3	Develop and strengthen research, development, adoption, and technological assimilation capabilities associated with information and communication technologies in municipal services.
PROSENER
PRONASE				T3	Identify and support institutional capacity building to expand technological, economic, environmental, and social research capabilities in relation to energy efficiency.
Strategy
PROSENER	T1	↑			Publish and update the National Inventory of Renewable Energy.
PETE	T1				Expand the National Inventory of Clean Energies (INEL, for its Spanish acronym) and the National Atlas of Areas with High Clean Energy Potential (AZEL, for its Spanish acronym) to incorporate new technologies and information from state and municipal governments.

Source: authors’ own elaboration.

shows some examples of T1 and T3 sequences. It is important to highlight that it is possible to find direct sequences between specific preconditions of either PETE or PRONASE to the final preconditions of the Strategy, skipping PROSENER (See Figure 11a). In such a case, this is also considered as a T3 sequence.

3.4. Systematization of Theory of Change Using Information Technologies

The case study presented in this paper used information technology (TI, for its Spanish acronym) to represent the ToC of PNTE-SEM. Therefore, a relational database in Microsoft Access©, as well as a model in Microsoft Power BI© using Data Analysis Expressions© (DAX) was created. This systematization using TI allows for an interactive and visual representation of all the elements of change while facilitating the understanding of the relationships between OPDs.

Hence, this work suggests the following general steps to implement the systematization process using TI:

• Firstly, LAs included in the Strategy, PROSENER, PETE, and PRONASE are analyzed to understand and extract the narrative and information that explain the relationships and pathways of change of PNTE-SEM.
• Secondly, an inventory of LAs is created in a relational database (Microsoft Access©) for each OPD, and then is structured in tables.
• Additionally, each LA contained in this inventory was sorted by the corresponding drivers that promote PNTE-SEM, namely, GCL, GDL, EE, RTYD, and SACC, and complemented with other relevant information, such as the typology used to classify the pathways of change, as well as with other relevant database fields.
• The resulting tables from the previous step are then imported to the model in Microsoft Power BI©. Thanks to the collection of functions, including those in DAX, such as NATURALINNERJOIN, NATURALLEFTOUTERJOIN, EXCEPT, and UNION, among others incorporated in this formula language, it is possible to automatically conduct an analysis of coherence, coordination, comprehensiveness, and consistency of LAs included in OPDs based on set theory’s operations and notations.
• Finally, the results are presented in different views and graphics to facilitate the interpretation of the key findings.

It is important to highlight that steps 1 to 3 are subject to the judgement of policy makers. If more complex interactions are evaluated, it is then necessary to analyze them exogenously by using other models and then integrate them as binary or composite relations. Therefore, these steps are crucial and must be considered when interpretating results.

4. Results of the Analysis on Coherence, Coordination, Comprehensiveness, and Consistency of Official Planning Documents

This section presents the main results of implementing the specific methodology and systematization process in the frame of the ToC described in the previous section in order to evaluate the OPDs of PNTE-SEM.

4.1. Analysis of Relationships

A total of 323 LAs contained in the OPDs were analyzed. As a result, 30% (98) exhibited strong intensity connections, 31% (99) had moderate intensity connections, and 39% (126) had weak intensity connections. In summary, 197 (61%) had at least one hierarchical relationship (connected with 2 or 3 documents), while 126 (39%) were not connected at all (See Figure 12).

4.1.1. Strong, Moderate, and Weak Intensity Connections

When the OPDs were individually analyzed, the results showed that PRONASE was the one with the highest share of strong intensity connections (46%), followed by PROSENER and PETE with 33% and 31%, respectively. In contrast, the Strategy had the lowest share of strong intensity connections (22%).

Regarding the moderate intensity connections, the Strategy had the highest share with 46%, followed by PETE with 27%, whereas PROSENER and PRONASE had the lowest shares, with 15% and 12%, respectively.

As for weak intensity connections, PROSENER was the one with the highest share with 52%, while PETE and PRONASE accounted for the same shares with 42%. The Strategy exhibited the lowest share, with 32% of all the LAs included in that document (see Figure 13).

4.1.2. Strong and Moderate Intensity Connections

When the LAs were analyzed by areas or drivers of PNTE-SEM (see Figure 14), GCL was the one with the highest share of strong and moderate intensity connections. Thus, PETE showed the highest share with 63%, followed by the Strategy and PROSENER, with 56% and 40%, respectively. In contrast, PRONASE had the lowest share, with 9%. It is worth mentioning that weak intensity connections were not included, as there is not a hierarchical relationship to evaluate with OPDs.

Regarding EE, the results pointed out PRONASE as the one with the highest share, with 85%, while PROSENER and the Strategy had shares of 28% and 27%, respectively. On the contrary, PETE was the one with the lowest share, with 3%.

As for GDL, the results placed it as the second most important area in PETE, with 27%, while it accounted for the third relevant area in the Strategy and PROSENER, with 15% and 12%, respectively. In contrast, no LAs devoted to GDL could be found in PRONASE.

Lastly, SACC and RTYD were the drivers covered to a lesser extent in the OPDs. PROSENER had shares of 8% and 12% for both drivers, while PETE had shares of 5% and 2%, respectively. The Strategy accounted for the lowest share, with 1% each, and PRONASE did not have any related LAs in the frame of SACC and exhibited a negligible share for RTYD.

4.1.3. Weak Intensity Connections

There is a significant number of policies with weak intensity connections (126), accounting for 39% of the total number of policies analyzed in this paper. Furthermore, as shown in Figure 15, when the LAs were grouped by areas or drivers of PNTE-SEM, the highest share was GCL, with 48%, followed by EE, with 30%. Other areas or drivers with the lowest share were GDL, RTYD, and SACC, with 10%, 7%, and 5%, respectively.

On the other hand, as depicted in Figure 16, when the results were analyzed by areas or drivers of PNTE-SEM and OPDs, it was found that EE was the one with the highest share of weak intensity connections, with 96% in PRONASE.

Similarly, GCL had the second highest share, with 72%, 63%, and 50% in PETE, PROSENER, and the Strategy, respectively.

The third area with the highest share of weak intensity connections was GDL, with 26% in the Strategy, although EE also exhibited important shares of weak intensity connections in the Strategy and PROSENER, with 20% and 15%, while this share was lower in PETE, with 7%.

Finally, SACC and RTYD were the areas with the lowest share of weak intensity connections, with 4% in PRONASE and the Strategy, respectively, although this share was higher for RTYD in PETE and PROSENER, with 14% and 11%, respectively. Similarly, this share was higher for SACC, with 7% and 11% in PETE and PROSENER, respectively.

4.2. Analysis of Sequences

As shown in

Table 4. Number of sequences in strong and moderate intensity policies by areas or drivers of Mexico’s energy transition policy in the electricity sector.

	Areas or Drivers
	GCL	GDL	EE	SACC	RTYD	Total
Strong intensity policies PETE -> PROSENER -> Strategy	15	17	11	1	0	44
Strong intensity policies PRONASE -> PROSENER -> Strategy	1	6	47	0	0	54
Moderate intensity policies PROSENER -> Strategy	37	1	8	0	0	46
Moderate intensity policies PETE -> PROSENER	2	0	0	1	0	3
Moderate intensity policies PRONASE -> PROSENER	2	0	0	0	3	5
Moderate intensity policies PETE -> Strategy	13	5	0	0	1	19
Moderate intensity policies PRONASE -> Strategy	0	2	9	0	0	11
Total	70	31	75	2	4	182

Source: authors’ own elaboration.

, a total of 182 sequences identified in this work were analyzed to evaluate their connections in accordance with the typology previously described in Table 1 and illustrated in Figure 6. The results show that policies with strong and moderate intensity connections were almost evenly distributed, with 98 and 84 sequences, respectively. As in the analysis of relationships, weak intensity connections are set aside, as they are only included in one OPD.

Furthermore, as previously illustrated in Figure 6, it is worth mentioning that one sequence in policies with strong intensity connections is made of two connections in three OPDs. Similarly, one sequence in policies with moderate intensity connections is made of one connection between two OPDs.

4.2.1. Strong Intensity Connections

As shown in

Table 5. Analysis of strong intensity connection policies in PETE -> Strategy -> PROSENER sequence.

PETE -> PROSENER	PROSENER -> Strategy	No. of Sequences PETE -> PROSENER -> Strategy	%
T1	T1	1	2
T1	T3	29	66
T2	T3	6	14
T3	T2	2	5
T3	T3	6	14
Total		44	100

Source: authors’ own elaboration.

, only one sequence with the desired typology (T1) was found out of 44 existing sequences from the specific preconditions of PETE to the initial preconditions of PROSENER, and from there to the final preconditions of the Strategy. On the other hand, the highest share of total sequences was T1 and T3, while the remaining sequences corresponded to T2 and T3 combinations. All in all, these results suggest a clear failure in the design of the OPDs in nearly all their sequences, of which a vast majority were related to GCL, GDL, and EE, while a rather low number corresponded to SACC (See Table 4).

Additionally,

Table 6. Analysis of strong intensity connection policies in PRONASE -> Strategy -> PROSENER sequence.

PRONASE -> PROSENER	PRONASE -> Strategy	No. of Sequences PRONASE -> PROSENER -> Strategy	%
T1	T2	1	2
T1	T3	40	74
T2	T3	12	22
T3	T3	1	2
Total		54	100

Source: authors’ own elaboration.

shows the existence of one sequence only (T1 and T2) out of 54 from the specific preconditions of PRONASE to the initial preconditions of PROSENER, and from there to the final preconditions of the Strategy. On the other hand, the highest share of total sequences was T1 and T3, while the remaining sequences corresponded to T2 and T3. Finally, there was only one incomplete T3 sequence.

These results also suggest inadequate coordination when designing OPDs, as there was only one sequence with the desired typology (T1). Moreover, there were several sequences from the final preconditions of the Strategy to the initial preconditions of PROSENER. This could also be interpreted as failures in coordination, since the policies were focused on the short-term rather than on the correct achievement of medium- to long-term conditions. Nearly all these inconsistencies were related to EE and a rather low number to GDL and GCL, while there were no sequences related to SACC and RTYD (See Table 4).

4.2.2. Moderate Intensity Connections

Table 7. Analysis of moderate intensity connection policies in Strategy -> PROSENER sequence.

Strategy -> PROSENER	No. of Sequences	%
T3	46	100
Total	46	100

Source: authors’ own elaboration.

shows that T3 sequences from the initial preconditions of PROSENER to the final preconditions of the Strategy had the highest participation in moderate intensity connections, with 46 out of 84. It is important to note that none of these sequences are well designed, as the initial preconditions of PROSENER are not connected to any specific preconditions of either PETE or PRONASE. Most of these sequences were related to GCL and EE and only one to GDL, whereas no sequences related to SACC and RTYD were detected (See Table 4).

On the other hand, the second highest share corresponded to 30 direct sequences from either PETE or PRONASE to the Strategy but skipped PROSENER (19 from PETE and 11 from PRONASE). This is also considered a failure in the planning design of the OPDs, since there is no connection between the specific preconditions and initial preconditions. Most of these failures corresponded to GCL and GDL and to GDL and EE for sequences originated from PETE and PRONASE, respectively. There was only one sequence contained in PETE related to RTYD (See Table 4).

Finally, as shown in

Table 8. Analysis of moderate intensity connection policies in PETE -> PROSENER sequence.

PETE -> PROSENER	No. of Sequences	%
T1	3	100
Total	3	100

Source: authors’ own elaboration.

and

Table 9. Analysis of moderate intensity connection policies in PRONASE -> PROSENER sequence.

PRONASE -> PROSENER	No. of Sequences	%
T1	1	20
T3	4	80
Total	5	100

Source: authors’ own elaboration.

, there was a rather low number of connections from the specific preconditions of either PRONASE or PETE to the initial preconditions of PROSENER, with 5 and 3 sequences, respectively (4 T1 and 4 T3). Most of these sequences were related to GCL and RTYD and only one to SACC (See Table 4).

5. Discussion

Due to the complex nature of hierarchical and causal relationships of policies and actions included in OPDs, it is necessary to assist policy makers with simplified and flexible approaches and analytical tools, such as the ones proposed in this work. Particularly, the ToC applied to the case study of the PNTE-SEM has shown its usefulness to identify areas of opportunity for a better design of OPDs in Mexico. Similarly, the proposed systematic evaluation using commercial software, such as Microsoft Access, and its exploitation in Microsoft Power BI facilitates the replication of this work in other countries that lead their planning efforts through strategies, special programs, and action plans. Since causal relationships have been modeled in this work based on policy makers’ experience, further work is still required to incorporate more complex interactions, for example, the water–energy nexus.

6. Conclusions

This article shows the importance of applying the ToC as a holistic and systemic approach for sequentially related policies and actions in OPDs that ideally are intended to achieve the short-, medium-, and long-term goals, as in this case study for PNTE-SEM.

The results derived from this work’s analysis suggest that OPDs have mostly privileged GCL as a driver for TE, as well as EE and GDL. In contrast, the policies and actions were covered to a lesser extent for RTYD and SACC, creating asymmetry not only between the expansion of GCL, GDL, and the strengthening and modernization of electrical grids, but also between energy and climate and environmental policy.

Consequently, PETE seems to be the most specific planning document designed for clean generation promotion, with 90% of all the policies and actions included in this document, while PRONASE appears to be the most specific planning document to promote EE, with 85% of all policies and actions included in the same document.

On the other hand, the obtained results also point out to an important disconnection between the policies and actions contained in OPDs. As a result, there are also important shares of weak intensity connection policies, especially for GCL and EE, which also suggest that the OPDs still lack a good design in terms of coherence, coordination, comprehensiveness, and consistency, according to the ToC.

Moreover, PROSENER and the Strategy are the most connected OPDs, hence, they set sectoral planning directives, although in most cases, not with the right sequence. Likewise, many policies and actions in these documents are connected to neither PETE nor PRONASE, which in turn reduces the possibility for them to be implemented in other specific programs.

Finally, there is a lack of coordination and consistency between policies and actions included in the OPDs as a result of failures in the desired sequence in accordance with the ToC, or, in other words, from the specific preconditions of either PETE or PRONASE to the initial preconditions of PROSENER and from there to the final preconditions of the Strategy. Particularly, this can be observed in connections involving PROSENER and the Strategy, as they barely exist in the correct and complete sequence between each other. Furthermore, there are important failures in the design of PETE and PRONASE since many sequences are connected directly to the Strategy, skipping PROSENER.