Proposed Reforms to the

Capex Approval Process for

Electric Cooperatives

Consultation Paper

Energy Regulatory Commission

December
2015

Copyright Castalia Limited. All rights reserved. Castalia is not liable for any loss caused by reliance on this
document. Castalia is a part of the worldwide Castalia Advisory Group.
Table of Contents
1 Introduction and Summary 1
1.1 Issues with Capex Approvals 1
1.2 The Proposed Approach 1
1.3 Consultation Process 2
2 Capex Approval Approach 3
2.1 Stage 1 Assessment of the overall proposal 3
2.1.1 Primary CriteriaConsistency with an approved
EICPM 4
2.1.2 Primary CriteriaCapex Spend Benchmark 5
2.1.3 Secondary Criteria for the Backlog 6
2.1.4 Stage One Assessment Process 6
2.1.5 Approval for Future Low Risk Applications 7
2.2 Stage 2 Assessment of projects in Medium Risk
Proposals 8
2.2.1 Project Assessment Components 8
2.2.2 Stage Two Assessment ProcessMedium Risk
Proposals 8
2.3 Stage 2 Assessment of projects in High Risk Proposals 9
2.3.1 Assessment of options 9
2.3.2 Assessment of demand projections 10
2.3.3 Assessment of supporting technical analysis 10
2.3.4 Assessment of economic evaluations 10
2.3.5 Stage Two Assessment ProcessHigh Risk
Proposals 10

Appendices
Appendix A Capex benchmark model ........................................................................12

Tables
Table A.1: Comparison of Benchmark Capex Model to Actual
Capex Approvals 12
Table A.2: Proposed Capex Model Regression Analysisbased on
approved Capex Proposal 13
Figures
Figure 2.1: NEA HarmonisationOption One 4
Figure 2.2: NEA HarmonisationOption Two 5
Figure 2.3: Future ApplicationsStage One Assessment 7
Figure 2.4: Future ApplicationsStage Two Assessment
Medium Risk 9
Figure 2.5: Future ApplicationsStage Two AssessmentHigh
Risk 11
 1 1 Introduction and Summary
2 We have been engaged by the Energy Regulatory Commission (ERC) to address the
3 process for the approval of the capital expenditure plans of the Electricity Cooperatives
4 (ECs), proposing measures for a more streamlined and efficient process for future
5 approvals.
6 1.1 Issues with Capex Approvals
7 There have been long delays in the approval of Capex plans and there is currently a large
8 backlog.
9 The reasons for these delays are many and include variable quality of submissions from
10 the ECs and the long public consultation and legal approval processes. However, the key
11 problem is that all proposals, regardless of risk are reviewed in the same complex and
12 detailed process.
13 The proposed changes to the processing of Capex plans will make the process more
14 efficient and streamlined as the level of review will be tailored to the level of risk. The
15 level of risk will be assessed as being both the probability that expenditure in the Capex
16 plan is unreasonable and/or the potential impact of such unreasonable expenditure being
17 passed on to consumers.
18 1.2 The Proposed Approach
19 The risk based approach has two stages:
20 Stage One is a proposal level evaluation that categorises proposals as low, medium or
21 high risk according to three objective criteria:
22 Primary CriteriaConsistency with an approved EICPM;
23 Primary CriteriaCapex Benchmark; and
24 Secondary CriteriaKPI and Compliance Benchmarks.
25 Those applications categorised as low risk will be approved at the proposal level.
26 Stage One is detailed in Section 2.1.
27 Stage Two takes those categorised as medium or high risk and subjects them to a
28 more detailed assessment at the project level using the following components:
29 options considered;
30 demand forecasts;
31 technical analysis; and
32 economic evaluation.
33 Stage Two for medium risk plans is detailed in Section 2.2 and for high risk plans in
34 Section 2.3.
35 The calculation of the RFSC Charge is linked to the ERC approval of an ECs Capex
36 plans. A separate consultation paper will be issued on the proposed changes to setting
37 the RFSC charge.
38

1
 1 1.3 Consultation Process
2 The aim of this consultation is to get feedback on our proposed changes to the Capex
3 approval process.
4 All interested parties are invited to submit their comments on this consultation paper on
5 or before 12th February, 2016. Electronic copies may be sent to tariffs.erc.gov.ph. This
6 consultation paper may be downloaded at the ERC website www.erc.gov.ph or may be
7 photocopied at cost at the ERC Main Office at the Docket Section, 18th Floor, Pacific
8 Center Building, San Miguel Avenue, Pasig City.
9

10

2
 1 2 Capex Approval Approach
2 For Capex approvals the proposal is for a risk based approach, categorising applications
3 as low, medium or high risk based on objective criteria. The criteria will include both
4 primary criteria and secondary criteria.
5 Those applications categorised as low risk will be approved at the proposal level.
6 Those categorised as medium or high risk will be subject to a more detailed assessment at
7 the project level.
8 This process applies to the review and approval of scheduled multi-year Capex proposals
9 for ECs. It assumes that ECs submit multi-year capex proposals in groups, with each
10 group to submit in separate years.
11 The suggested approach has two stages:
12 Stage 1 Risk assessment and approval for low risk proposals
13 Stage 2 Project assessment for medium and high risk proposals
14 Stage 1 assesses whether there is a high risk that the Capex proposals exceed the
15 reasonable needs of the EC. Proposals that are assessed to be low risk are approved at
16 this stage. Other proposals are classified as either medium or high risk and subject to a
17 more detailed second stage analysis.
18 Stage 1 places considerable reliance on the consistency with the program of works in an
19 EICPM approved by NEA. This will require greater coordination between NEA and
20 ERC. Stage 2 assesses the individual projects within the overall proposal in terms of the
21 option analysis, demand forecasts, technical analysis, and economic analysis. While the
22 structure of the process is the same for medium and high risk projects, the extent of
23 scrutiny is greater for high risk projects. The scrutiny of projects in high risk proposals
24 assesses the details of the supporting analysis and is similar to the current level of analysis
25 of Capex proposals.
26 The primary objective of a review is to reduce the risk of overspending. Furthermore, the
27 review process may delay projects and increase the risk of underspending. Hence, the
28 reviews should be focussed on higher risk proposals. The consequences of over-spending
29 and under-spending on Capex are the same across the high, medium and low risk
30 projects but the likelihoods are different. In the absence of a review the likelihood of
31 overspending is greatest for the high risk proposals and lowest for the low risk projects.
32 Hence, substantially greater scrutiny is applied to high risk proposals.
33 The assessment criteria fit closely with the requirements of the Planning Manual. Hence
34 adoption of this approach would not weaken the ERCs commitment to the proper
35 implementation of the planning manual, but it would make the testing of its application
36 through the Capex approval process more streamlined. Some modification of the ERCs
37 existing rule on the review of Capex programs would be required but this would not be
38 substantial.
39 2.1 Stage 1 Assessment of the overall proposal
40 The Stage 1 assessment has three elements:
41 Primary CriteriaConsistency with an approved EICPM;
42 Primary CriteriaCapex Benchmark; and
43 Secondary CriteriaKPI and Compliance Benchmarks.

3
 1 2.1.1 Primary CriteriaConsistency with an approved EICPM
2 The first criteria for assessing Capex proposals is whether they are consistent with the
3 latest approved EICPM. This requires that each of the projects submitted in the
4 proposed capex plan is part of the program of works in the EICPM so that ERC can
5 place greater reliance on the approval processes of NEA.
6 To be effective this requires the coordination of the submission of Capex proposals to
7 NEA and ERC.
8 Capex proposals that meet this criteria will be assessed as low risk if they meet all the
9 other criteria. Those proposals that do not meet this criteria will be assessed as high risk.
10 Given the proposed coordination of the submission of Capex proposals it is expected
11 that most ECs will meet this criteria if their Capex planning process is well managed and
12 they meet the procedural requirements.
13 Currently ECs submit Capex proposals to NEA as part of the EICPM every second year
14 in January and these are approved by the end of March. ERC does not have the
15 resources to process applications from half the ECs concurrently. Hence, it is proposed
16 that NEA and ERC harmonise the applications to a three year cycle.
17 The two options for synchronisation are shown in Figure 2.1 and Figure 2.2:
18 Figure 2.1: NEA HarmonisationOption One

19
20

4
 1 Figure 2.2: NEA HarmonisationOption Two

2
3
4 On balance Option 1 is considered likely to be more feasible and a better compromise
5 between the needs of NEA and ERC.
6 2.1.2 Primary CriteriaCapex Spend Benchmark
7 The proposed Capex benchmarkthe primary criteriais an Asset Based Capex Model.
8 The benchmark cost model is tailored to the specific cost drivers of each EC. In
9 principle a cost function will:
10 Retain the benefit of low cost and low effort regulation;
11 Set a benchmark that is tailored to each EC; and
12 Reflect the specific characteristics of the ECs in evaluating Capex proposals.
13 We have tested a number of different cost models using known drivers of network
14 capital costs. All cost models were tested against recently approved Capex plans to assess
15 the predicative power of different models, the significance of cost drivers and to ensure
16 each coefficient has the correct sign.
17 An initial model to estimate reasonable capex needs has been developed and shows that
18 physical measures of the core network assets (network length and substation capacity) are
19 the best predictors of Capex needs.
20 A non-linear model is preferred to a linear model, given that network costs are non-linear
21 and because a log-log model is the best fit to the diminishing returns from economies of
22 scale. The proposed Capex Model will use two core network attributes to calculate a
23 benchmark for each EC:
24

5
 1 Total network circuit length in km, and
2 Substation capacity in MVA.
3 Based on regression analysis, the benchmark Capex spend will be calculated as follows:
4

[ ( ) ( ]

5
6 Growth and asset age measures were not statistically significant in our initial analysis but
7 may be included in the future to improve the overall performance of the model. The
8 Capex model will be continually developed and refined over time in a transparent process
9 so that ECs understand how it is used in the risk categorisation of proposals. A detailed
10 review of the Capex benchmark model is provided in Appendix A.
11 The aim of the model is to provide a guide to the typical level of Capex that would be
12 reasonable. Inevitably there will be cycles in Capex so that exceeding the benchmark does
13 not mean that the program is necessarily too high. But it will provide an indicator of
14 where further analysis (Stage Two) is warranted.
15 If the Capex proposals are higher than the benchmark they will be assessed as medium
16 risk as long as they meet both secondary criteria. Otherwise they will be assessed as high
17 risk. We expect that typically half or more of the proposals would be below the
18 benchmark.
19
20 2.1.3 Secondary Criteria for the Backlog
21 We suggest that the simple and objective secondary criteria, for future applications be:
22 Are KPIs met? (e.g. submission of data and compliance returns); and
23 Quality of justifications (i.e. assessed to be better than average).
24 Assessment of quality of justifications
25 Inevitably this will be some degree of subjectivity and the challenge will be to ensure that
26 the assessment is undertaken against clear criteria and is consistent. The key point is that
27 this is a relative assessmentin principle half of the proposals will be assessed as having
28 better supporting material than average and half having poorer supporting material.
29 There are two key steps to facilitate this:
30 Establishment of criteria and standardsthis can draw upon the
31 development of criteria and standards for the assessment of the supporting
32 analysis for existing applications; and
33 Participation in NEA regional workshops with ECsas part of the
34 NEAs review process, NEA staff hold regional workshops where they test
35 the underlying analysis for the ECs proposals. ERC staff would participate in
36 these meetings as observers. This will be efficient means of assessing the
37 comparative quality of the underlying analysis.
38 2.1.4 Stage One Assessment Process
39 The recommended Stage One assessment process flowchart for future applications is
40 shown in Figure 2.3.

6
 1 Figure 2.3: Future ApplicationsStage One Assessment

2
3
4 2.1.5 Approval for Future Low Risk Applications
5 If they meet the primary and secondary criteria, future applications would fall under the
6 low risk category, thus they will be approved at the proposal level with no further
7 scrutiny.
8 This is because they have been assessed as being consistent with an approved EICPM,
9 below the Capex benchmark for a reasonable level of expenditure and have been judged
10 to have above average quality of justification and analysis.
11 A compliance check with the EICPM is an important measure for reducing ERCs
12 current burden and to accelerate the approval process. This is because, it incentives the
13 ECs to submit fully compliant Capex proposal submissions, and avoids unnecessary
14 interactions in gathering missing information. ECs that do not conform to the EICPM,
15 would face a more rigorous review of their Capex submission and delays in receiving
16 ERC approval.
17

7
 1 2.2 Stage 2 Assessment of projects in Medium Risk Proposals
2 For those Capex plans that are categorised as medium risk, each of the components of
3 the underlying analysis is tested against objective, transparent criteriathrough a series
4 of Yes or No questions. This means that the extent of judgement required is
5 minimised and detailed analysis of the supporting material is not required. This will
6 facilitate prompt review and decision-making and standardised decision documents. The
7 questions focus on the key requirements of the EC Distribution Utility Planning Manual
8 (ECDUPM). ECs whose Capex programs comply with the requirements of the
9 ECDUPM should be able to readily demonstrate compliance with the assessment criteria
10 for medium risk proposals.
11 The key difference from the Stage 1 assessment is that the Stage 2 assessment happens at
12 the project level.
13 2.2.1 Project Assessment Components
14 The components to be assessed are:
15 options considered;
16 demand forecasts;
17 technical analysis; and
18 economic evaluation.
19 A key part of the assessment will be the reliance on NEAs assessment of the demand
20 forecasts. This will be made possible by coordinating the submission of Capex proposals
21 with the EICPM as discussed in Section 2.1.1. NEA places considerable emphasis on the
22 assessment of the demand forecasts and ensuring the forecasts are consistent with the
23 requirements of the planning manual. Hence, it is proposed to draw upon this rather
24 than duplicating the effort of NEA.
25 2.2.2 Stage Two Assessment ProcessMedium Risk Proposals
26 The recommended Stage Two project by project assessment for future applications that
27 are deemed as medium risk is shown in Figure 2.4.

8
 1 Figure 2.4: Future ApplicationsStage Two AssessmentMedium Risk

Future Standard Applications Medium Risk

Individual
Projects

No Test: the proposal

Are options
No includes an evaluation
evaluated?
of options

Yes

Test: Are the forecasts

No Is demand
consistent with latest
No forecast
forecasts accepted by
reasonable?
NEA (as part of the
eICPM).
Yes

No Is technical
Test: Is the analysis
No analysis consistent with
adequate? demand forecasts and
standards?

Yes

No Are economic
No evaluations
Test: Is an economic
provided?
evaluation provided?

Yes

Reject Approve
2
3
4 2.3 Stage 2 Assessment of projects in High Risk Proposals
5 This process has the same structure as the assessment of projects in medium risk
6 proposals but requires a more detailed qualitative assessment of the components, other
7 than the demand forecast component. The level of analysis may also vary according to
8 the size of the project. That is, the larger or more unusual projects should be subject to
9 closer scrutiny than smaller or routine projects.
10 2.3.1 Assessment of options
11 Unlike the medium risk assessment, the ERC will consider not just whether options are
12 considered but whether the options are appropriate. That is, do the options cover a
13 set of alternatives that are likely to meet the identified needs and include the best
14 available options? This does not require the reviewing engineer to undertake the option
15 analysis but if it is determined that the options are deficient it will be necessary to show

9
 1 in what way they are and identify or excluded options that could have been considered.
2 In practice this is likely to be the most critical part of the evaluation of the projects.
3 2.3.2 Assessment of demand projections
4 Reliance is placed on the NEAs analysis of the demand projections. Further analysis will
5 only be required if the demand projections are not consistent with the approved EICPM.
6 This is unlikely, given the proposed coordination with the EICPM.
7 2.3.3 Assessment of supporting technical analysis
8 The focus of the assessment of the technical analysis remains on its consistency with the
9 demand forecasts and network planning standards. In addition, the engineer should test
10 whether the analysis clearly supports the proposed option over the alternatives. If
11 inconsistencies are identified that raise questions about the underlying models used these
12 should be investigated but these models would not normally be reviewed in detail.
13 2.3.4 Assessment of economic evaluations
14 For projects in high risk proposals the reviewing engineer is required to assess whether
15 the economic evaluation is adequate. The key questions in assessing this will be:
16 Whether the methodology used is consistent with the requirements of the
17 planning manual;
18 Whether the analysis is consistent with the demand forecasts, quality of
19 service, and network planning standards;
20 Whether the assumed costs of the proposal and options considered are
21 reasonable;
22 Whether the other assumptions that may be used (e.g. discount rate, project
23 life, and value of lost load) are reasonable; and
24 Whether the analysis supports the proposal over the alternative options.
25 This focuses the assessment on the methodology and assumptions. The reviewing
26 engineer would not normally undertake a detailed review of the underlying models unless
27 the results raise significant questions about the modelling.
28 2.3.5 Stage Two Assessment ProcessHigh Risk Proposals
29 The recommended Stage Two project by project assessment flowchart for future
30 applications assessed as high risk is shown in Figure 2.5.

10
1 Figure 2.5: Future ApplicationsStage Two AssessmentHigh Risk

Future Standard Applications High Risk

Individual
Projects

No Test: Are appropriate

Are options
No alternatives specified
evaluated?
and evaluated?

Yes

Test: Are the forecasts

consistent with latest
No Are the forecasts accepted by
No forecasting NEA (as part of the
models valid? eICPM). If not, why?
Are the models valid?

Yes

Test: Is the analysis

No Is technical consistent with
No analysis demand forecasts and
adequate? standards? Does it
support the selected
option?
Yes

Test: Review the

adequacy of the
No Are economic
evaluation. Are the
No evaluations
methodology and
valid?
assumptions valid? Do
the results show the
selected option is the
Yes best?

Reject Approve
2
3
4
5

11
 1 Appendix A: Capex Benchmark Model
2 Table A.1: Comparison of Benchmark Capex Model to Actual Capex Approvals
Total Annual Annual
Regulatory
# EC Acronym Capex Benchmark Variance
Period
Approved Capex
1 AKELCO 2011-2014 41,698,425 63,210,063 51.6%
2 ASELCO 2011-2013 20,067,089 53,758,919 167.9%
3 BOHECO II 2011-2014 45,179,360 41,247,827 -8.7%
4 CANORECO 2010-2014 31,281,111 41,125,910 31.5%
5 CEBECO I 2014-2018 52,697,961 60,777,917 15.3%
6 CELCO 2011-2015 3,516,308 2,472,516 -29.7%
7 CENPELCO 2011-2014 139,503,425 84,564,144 -39.4%
8 COTELCO 2012-2015 104,213,816 91,358,854 -12.3%
9 DIELCO 2012-2016 1,871,743 4,716,841 152.0%
10 FICELCO 2011-2015 12,011,318 13,048,544 8.6%
11 FLECO 2011-2014 52,867,089 13,067,593 -75.3%
12 GUIMELCO 2011-2015 16,874,809 9,627,116 -42.9%
13 ILECO II 2012-2013 59,421,046 40,551,718 -31.8%
14 ISELCO II 2011-2015 67,846,860 34,044,893 -49.8%
15 MORESCO I 2011-2014 209,093,055 73,427,510 -64.9%
16 ORMECO 2012-2016 73,230,715 87,372,300 19.3%
17 PRESCO 2010-2013 4,151,599 5,029,895 21.2%
18 PROSIELCO 2011-2015 4,281,836 4,898,268 14.4%
19 SORECO II 2011-2011 6,669,964 20,200,562 202.9%
20 ZAMCELCO 2009-2013 19,111,457 61,102,471 219.7%
21 ZAMSURECO II 2011-2014 62,695,734 39,879,596 -36.4%

3
4
5
6
7
8
9
10
11
12
13

12
 1
2 Table A.2: Proposed Capex Model Regression Analysisbased on approved
3 Capex Proposal
Model 8Coefficiencts match and all variabale are significant

Regression Statistics
R 0.8535
R Square 0.72846
Adjusted R Square 0.69829
Standard Error 0.31309
Total Number Of Cases 21
LOG (ANNUAL TOTAL ) = 4.0170 + 0.7884 * Log(Ckt length) + 0.5006 * Log(SS)

ANOVA
d.f. SS MS F p-level
Regression 2. 4.73361 2.3668 24.14411 0.00001
Residual 18. 1.76451 0.09803
Total 20. 6.49811

Coefficients Standard Error LCL UCL t Stat p-level H0 (10%) rejected?

Intercept 4.017 0.8915 2.47109 5.56291 4.5059 0.00027 Yes
Log(Ckt length) 0.78844 0.3468 0.18706 1.38982 2.27343 0.03548 Yes
4 Log(SS) 0.50061 0.25437 0.05951 0.94171 1.96802 0.06467 Yes

5
6
7
8
9
10

13
T: +1 (202) 466-6790
F: +1 (202) 466-6797
1747 Pennsylvania Avenue
NW 12th Floor
WASHINGTON DC 20006
United States of America

T: +1 (646) 632-3770
F: +1 (212) 682-0278
200 Park Avenue Suite 1744
NEW YORK NY 10166
United States of America

T: +61 (2) 9231 6862

F: +61 (2) 9231 3847
Level 1 Kyle House
27 31 Macquarie Place
SYDNEY NSW 2000
Australia

T: +64 (4) 913 2800

F: +64 (4) 913 2808
Level 2, 88 The Terrace
PO Box 10-225
WELLINGTON 6143
New Zealand

T: +57 (1) 646 6626

F: +57 (1) 646 6850
Calle 100 No. 7-33
Torre 1, Piso 14
BOGOT
Colombia

T: +33 (1) 45 27 24 55
F: +33 (1) 45 20 17 69
7 Rue Claude Chahu
PARIS 75116
France
------------- www.castalia-advisors.com