October 2011

1. Energy - Utilities October 2011 Measuring productivity at utilitiesReport prepared for Hydro OneIntervener meeting; October 19, 2011

2. Contents • Background • Executive summary • Industry measurement of productivity • By regulatory commissions • At utilities • Perspectives on productivity metrics • Activity cost analysis to choose metrics with the biggest impact • Principal cost driver analysis • Specific metric recommendations • Implementation considerations

3. Background • “In its December 23, 2010 Decision approving Transmission Revenue Requirements for 2011 and 2012, the Ontario Energy Board provided direction and other expectations for further information on compensation and efficiency comparisons. • The Board directed “Hydro One to revisit its compensation cost benchmarking study [the Mercer study] in an effort to more appropriately compare compensation costs to those of other regulated transmission and/or distribution utilities in North America.” • Toward that end, the Board directed "Hydro One to consult with stakeholders about how the Mercer study should be updated and expanded to produce such analyses”. • The Board went on to describe its expectation that Hydro One “be in a position to provide more robust evidence on initiatives to achieve a level of cost per employee closer to market value at its next transmission rate case. The Board will expect compensation increase to be matched with demonstrated productivity gains”. Extract from Hydro One RFP # SCO-1000152789, March 2nd 2011 • To satisfy all aspects of the boards request contained in the RFP Oliver Wyman was engaged alongside Mercer. • Mercer to update the compensation benchmarking study in 2011 and will separately report changes in relative compensation levels • Oliver Wyman to provide a perspectives on industry best practices for productivity measurement

4. Outline of approachInternal and external research was conducted to build a set of recommendations around how Hydro One could measure productivity. Research Recommendation Implementation • Contacted commissions • Contacted 9 US and 8 Canadian commissions about how they measure productivity. Outline implementation considerations • Surveyed utilities • Contacted 11 US and 20 Canadian utilities to establish how they measure productivity internally • 13 provided information • Next steps • Future technology • Implementation costs • Reporting and analysis • Assembled potential metrics • 10+ meetings held to discuss productivity with VPs and SVPs • Aggregated OW knowledge on best practices for balanced scorecard creation Provided set of metrics for consideration Performed cost analysis of provided financial data - Identified largest activities to target with metrics Reviewed the data capabilities of Hydro One - Systems expert interviews and viewed reporting tools

5. Executive summary • Oliver Wyman conducted a broad market survey of US and Canadian utilities and contacted many regulators directly to assess how common productivity measures were used. In total we made over 350 contacts to: • In Canada: 20 utilities and 8 Commissions • In the US: 11 utilities and 9 Commissions • This research found: • No commission was found to routinely measure productivity directly. Instead commissions focused on ‘outcome’ metrics of overall Service Quality Metrics (SQM) and total costs. In the majority of cases, the commissions directed us to contact utilities directly as the management of productivity was considered part of their internal management. • Most utilities did look at productivity metrics internally as part of a balanced scorecard to support understanding of trends of the SQM and Cost metrics reported to commissions. The productivity metrics found suggest that none of the participants track productivity across all business functions. • After analyzing Hydro One’s major costs and interviewing many of their senior staff, a number of metrics have been suggested as candidates to measure productivity, which account for 23% of total O&M and Capex costs.

6. Contents • Background • Executive summary • Industry measurement of productivity • By regulatory commissions • At utilities • Perspectives on productivity metrics • Activity cost analysis to choose metrics with the biggest impact • Principal cost driver analysis • Specific metric recommendations • Implementation considerations

7. Primary research at US and Canadian RegulatorsA majority of the regulators examined, measured total costs and service quality metrics instead of productivity metrics • 17 Regulators across the US and Canada were requested to provide any methodologies they had for measuring productivity • Contact was made by a combination of calls, e-mails and requests for information • A review was also performed of publicly filed documents such as rate cases and other dockets • Most regulators collected service quality metrics with a significant number of cost metrics. • No regulator regularly measured any productivity metrics • Nova Scotia Utilities and Review Board measured one productivity metric (Calls handled per agent per day) via an ad hoc study • Various additional studies were found during the research • Ad hoc studies: • Operational review of Nova Scotia Power, consultant benchmarked NSPI’s corporate services functions finding it was average to good performers in comparison with the panel (primarily cost benchmarks) • Kaiser Associates operational review of Nova Scotia Power benchmarked OM&G expenses determining NSPI performed at a lower cost than other utilities when normalized for size • CAMPUT review of benchmarking practices for regulators in Canada provided a background on the difficulties in benchmarking (inconsistent data, definition inconsistencies), as well as high level steps which must be taken before benchmarking can be adopted into regulatory proceedings (building support and consensus, data collection and summary, analysis and evolution) • Regular studies: • PSEG internal benchmarking survey which benchmarked safety, service quality, cost and reliability across a panel of electric utilities

8. Contents • Background • Executive summary • Industry measurement of productivity • By regulatory commissions • At utilities • Perspectives on productivity metrics • Activity cost analysis to choose metrics with the biggest impact • Principal cost driver analysis • Specific metric recommendations • Implementation considerations

9. Overview of methodology for utilities productivity surveyA number of utilities were surveyed to understand which metrics they collect internally • The purpose of the survey is to gain insight into the different ways in which utilities measure their performance internally in terms of service quality, cost and productivity • The survey required participants to provide their cost, productivity and service quality metrics and for each metric: • How long they had been measured for • If they were reported to regulators • If it was part of penalty • If they were benchmarked Respondents Survey panel recipients Canada United States Note: For a copy of the survey, please see the appendix; PSEG information was extracted from public sources

10. Summary of findings from productivity surveyPerformance was measured differently between each utility which participated in the survey, with most focusing on service quality. Overall Trends • Wide disparity in internal performance measurement, each utility defines productivity, service quality and cost metrics differently. • Generally each utility is unsure of what percentage of the total cost is being covered by the performance metrics submitted Category Category findings Metrics collected per utility Service Quality • Service quality metrics are reported to regulators more than cost or productivity • Common metrics: SAIDI, SAIFI, CAIDI, OSHA rates Median Max Min Total 25 176 4 478 Productivity • Most productivity metrics were also cost related on a unit basis • Cost / KM of line trimmed • Productivity metrics are not generally benchmarked, part of a penalty or reported to regulators Median Max Min Total 4 59 0 114 Cost • Most metrics detail overall spend in business categories, many have normalizing factors. E.g. Distribution O&M/customer • 22% of all cost metrics reported internally are part of a benchmarking effort Median Max Min Total 6 89 1 213

11. Service quality 34common metrics • System avg. interruption frequency index (SAIFI) • Customer avg. interruption disruption index (CAIDI) • % of calls answered in 30s or less • System avg. interruption duration index (SAIDI) • % of Calls abandoned • % of Meters read • % In-service appointments met • Customers experiencing multiple interruptions (CEMI) • Bill accuracy rate • Average speed of answer • Occupational Safety and Health Administration Incidence Rate • Momentary avg. interruption frequency index (MAIFI) • Emergency response time • SAIFI – Distribution Only • # of Off-cycle meter reads/month • SAIDI – Distribution Only • Occupational Safety and Health Administration Severity Rate • # of Postーfinal adjustment mechanism processed per month • New service installation factor • # of Sites billed/month • # of Sites not billed/month • DTE cases per 1000 customers • Damages per 1000 elect. Locate requests • Customer satisfaction – overall • Customer experience long interruption duration (CELID) • CAIDI – Distribution Only • CAIDI – Storm • Average number of energizations per month • Average number of de-energizations/month • Average System Availability Index (ASAI) • % of Meters not read within 6 months • % of Completed off-cycle meter reads >5 days • % of Calls answered in under 20s • Vehicle accident frequency rate Common metricsMetrics collected in the surveying process which were measured by at least two utilities. Cost 3 common metrics • Net income • Net Income from operations • OM&A Costs per customer Productivity 6 common metrics • Turnover • Cost per call • Meter reads per FTE • Lost time accident rate • First call resolution rate • Average time per calｌ

12. Contents • Background • Executive summary • Industry measurement of productivity • By regulatory commissions • At utilities • Perspectives on productivity metrics • Activity cost analysis to choose metrics with the biggest impact • Principal cost driver analysis • Specific metric recommendations • Implementation considerations

13. Criteria for choosing a set of metricsMost utilities select metrics to examine using some criteria that best fits with their business needs. A metric may need to be used in conjunction with other metrics to meet all the criteria above

14. Even within the same utility over time, job conditions change across the territory and in different years • It is important to obtain context of the metrics before making decisions • E.g., it is considerably more difficult to place a pole in rock than soil; work mix can mask productivity changes • Savings from improvements in working methods, which eliminate work may not be captured • For example, AMR has reduced the need for personnel to read meters, which is an improvement in productivity • There are potentially large costs associated with data collection, retention and analysis. • For example, collecting time associated with a specific task and associated conditions either requires mobile/handheld field systems or extensive overhead to collect the information at a detailed enough level. Considerations of productivity metric collectionThere are several considerations when using productivity metrics to make decisions about the performance of the operations. Balanced approach required Difficulty to obtain like for like comparisons Metrics could not capture all productivity improvements Cost of metric collection • Examining productivity in isolation can cause misunderstanding and unintended behaviors • Quality, safety, and long-term performance should be measured and examined in the same context. • e.g., encouraging faster job-time could lead to assets that malfunction and lead to lower reliability Despite the challenges utilities continue to look for ways to measure productivity in a way that improves their business • Most utilities use a balanced set of metrics to obtain the clearest picture of performance • Analysis teams place results into context of business cycles and external influences (e.g. weather) • Trends in headline metrics are explained by the underlying supporting metrics • Productivity metrics become integrated into the overall management of a utility aligned with corporate goals • Understand and explain the drivers behind changing costs, internally and to regulators • Target corporate initiatives at poorly performing areas • Utilities leverage advanced IT systems such as mobile tracking devices to produce detailed productivity metrics without creating large indirect costs

15. Contents • Background • Executive summary • Industry measurement of productivity • By regulatory commissions • At utilities • Perspectives on productivity metrics • Activity cost analysis to choose metrics with the biggest impact • Principal cost driver analysis • Specific metric recommendations • Implementation considerations

16. Costs of materials from all projects excluded because costs are driven by base commodity prices. Costs outside the top 20 projects of main functional areas. Not considered in first cut of metrics. Costs from top 20 projects in each category. These costs are broken out further in the next few slides. Analysis roadmap for choosing a suite of appropriate metrics Understanding the breakdown of spend on resource is the first step to determine the areas to measure. Total Spend 33% 26% 18% 12% 11% 100% 14% 24% 18% 31% 80% 16% 31% 60% 100% 79% 40% 69% 63% 45% 20% Top 20 projects in the 4 areas account for 64% of cost 0% 20% 40% 60% 80% 100% Trans. – OM&A. Trans. - Capital Dist. - Capital Dist. – OM&A. Other TC DC DM TM

17. Legend Relevant Metric Need more info. Not measurable Example: Distribution OM&A project metricsEight of the biggest twenty distribution projects have suitable metrics to measure part of their costs DM Cumulative cost of projects Percent of Total Cost Top 20 projects cover 79% of distribution maintenance cost Projects ranked by largest cost Note: Cost does not include materials or projects with negative or zero costs.

18. Summary of recommended metricsTwenty five selected metrics account for approximately twenty two percent of total costs. • Aggregating the metric choices from the previous pages, shows a good coverage of total cost • 22% of costs covered using 25 main productivity metrics. • Some metrics cover multiple activities across different segments (TC, T OM&A, DC, DOM&A). • Some further subdivision of these metrics may be required to allow better comparisons. • e.g. cost per pole could be sub divided into cost per pole per ground type. • Note that estimations of cost coverage was based on project titles. • Further validation with business would be required to confirm the assumptions made. • Large amount of projects could not be understood from titles well enough to suggest metrics.

19. Contents • Background • Executive summary • Industry measurement of productivity • By regulatory commissions • At utilities • Perspectives on productivity metrics • Activity cost analysis to choose metrics with the biggest impact • Overview of methodology • Principal cost driver analysis • Specific metric recommendations • Implementation considerations

20. Implementation considerationsHydro One will require a detailed plan to develop a set of productivity metrics that are integrated and aligned with the overall corporate scorecard and direction.