Demand Side Management in California

National Economic Review

National Institute of Economic and Industry Research

No. 60               December 2006

The National Economic Review is published four times each year under the auspices of the Institute’s Academic Board.

The Review contains articles on economic and social issues relevant to Australia. While the Institute endeavours to provide reliable forecasts and believes material published in the Review is accurate it will not be liable for any claim by any party acting on such information.

Editor: Dr A. Scott Lowson

© National Institute of Economic and Industry Research

This journal is subject to copyright. Apart from such purposes as study, research, criticism or review as provided by the Copyright Act no part may be reproduced without the consent in writing of the Institute.

ISSN 0813-9474

Demand side management in California: current and proposed measures

Graham Armstrong, NIEIR

Abstract

Although definitions vary, demand side management (DSM), demand management (DM) and demand response (DR) measures generally encompass energy efficient improvement, load shifting and peak load control. Over the past five years, increasing peak load demands and regional supply shortfalls (due to one or a combination of inadequate inter-connections, generator capacity, unexpected summer load peaks) have focused DSM/DR/DM efforts on peak load control of air conditioning equipment.

In Australia air conditioning loads are increasing at a rate of about 50 per cent above overall load growth. Although there has been increasing interest over the past five years in DSM to address this peak load growth, there have been few actions beyond analysis and discussion of the issue. Peak load growth has been met by supply augmentation.

On the other hand in California, where electricity prices soared and supply shortfalls were experienced in 2000, a range of measures has been introduced.1 Today, California is almost certainly the jurisdiction with the most comprehensive array of DSM/DM/DR measures. These measures are mainly designed (often with overall government direction by the State’s Government) and delivered by energy utilities operating in the State.

Graham Armstrong believes that the United States experiences with measures for addressing peak loads are useful when considering the situation in Victoria and Australia in general – with the important qualification that policy design must be based on our particular circumstances and provides a preliminary program design for consideration and analysis.

Introduction

California can in some ways be viewed as a stand-alone nation state which has the fifth largest economy in the world. The Californian electricity demand requires a capacity of nearly 55,000 MW (about 25 per cent imported): this compares with total Australian generation capacity of about 50,000 MW (Victoria 8,000 MW). 2 Accordingly, California is a very significant global entity in the energy field.

California – A Nation State

  • Population of 34 million in 2002, 41 million by 2010.
  • 5th largest economy in the world.
  • 5th largest consumer of energy in the world.
  • 2nd largest consumer of gasoline and diesel – only the total United States uses more.
  • Lowest US per capita electricity consumption.
  • 1.5 per cent of world’s greenhouse gas emissions but low per capita emissions.

Source:   California Climate Change Programs: An Overview, Conference of the Producers, The Hague, 12 May 2003 presented by James D. Boyd, Californian Energy Commission.

In 2004 the State electricity usage was about 265,000 gigawatt hours of electricity per year. Consumption is growing at 2 per cent annually. Over the 1994-2004 period, between 29 per cent and 42 per cent of California’s in-state generation used natural gas. Another 10 -20 per cent was provided by hydroelectric power that is subject to significant annual variations. Almost one third of California’s entire in-state generation base is over 40 years old. California’s transmission system is also ageing. While in-state generation resources provide the majority (average annual of about 75 per cent) of California’s power, California is part of a larger system that includes all of western North America. Fifteen to thirty per cent of state-wide electricity demand is imported from sources outside State borders.

Peak electricity demands occur on hot summer days. California’s highest peak demand was 52,863 megawatts which occurred on 10 July 2002. On average peak demand is growing at about 2.4 per cent per year, requiring the equivalent of about three new 400 MW peaking power plants per year. Residential and commercial air conditioning represent at least 30 per cent of summer peak electricity loads.

California’s demand for natural gas also is increasing. Currently the State uses 2 trillion cubic feet (2,100 PJ, Victoria approximately 250 PJ) of natural gas per year. Historically the primary use of this fuel was for space heating in homes and businesses. Electricity generation’s dependence on relatively clean burning natural gas now means that California’s annual natural gas use by power plants is expected to increase. Overall, natural gas use is growing by 1.6 per cent per year. Eighty five per cent of natural gas consumed in California is supplied by pipelines from sources outside the State.

Californian initiatives in DSM/DR/DM have evolved in three fairly distinct phases over the past 30 years.

In the first phase, extending from the mid 1970s to around 1990, the emphasis, led by utilities such as Pacific General Electric (PGE), was on energy efficiency in an integrated resource planning (IRP) framework, in which the costs of reducing energy demand were compared with the costs of expanding supply. In this phase measures focused on energy efficiency with some attention to load control.

In the second phase, extending into the 1990s, less attention was paid to DM/DSM/DR as supply pressures (costs, levels) eased: a situation common around the world. Environmental concerns increased, particularly urban air quality and greenhouse, but more attention was paid to transport rather than stationary energy. DSM funding, focused on energy efficiency improvement (EEI) varied considerably in the period as regulatory wrangles remained unresolved.

The third phase, commencing in 2000, was precipitated by the electricity supply disruption and soaring wholesale prices. Since then DSM/DR/DM measures (both voluntary and the use of incentives) have been vigorously pursued with substantial public spending. The 2001 summer peak, weather and growth adjusted, was 10 per cent below the 2000 peak. The immediate response to the 2000 events was to install emergency peaking plants and to engage in a publicity campaign and incentive measures (lower tariffs for reducing demand below the previous year) to curtail demands. Rebates for the purchase of higher efficiency products were also tried to curb power consumption in 2000-01, but this approach was judged to be relatively ineffective as take-up was low and wholesale electricity prices fluctuated from one hour to the next, but retail prices did not.

Program funding has mainly been based on a combination of State funds provided on measured energy savings and utility funding, but in 2000 -01 the Californian Energy Commission (CEC) was appropriated an additional $ 380 million from special taxpayer funds for a range of DSM programs.

Recent developments

Although rebates continue, the policy focus has shifted to the potential use of time-of-use (interval) meters, which could be used with time -of-use (t-o-u) pricing (dynamic pricing in Californian terms, which includes consideration of real time pricing, RTP, covering price changes as wholesale prices change).

Backed by data from the t-o-u meters, rates can be adjusted according to several market variables, including demand, supply, wholesale prices and individual use. The State, with the major utilities, conducted a test to gauge customer response to variable pricing. About 2,500 small scale users across the State were given t-o-u meters and put on different pricing plans. In one plan, consumers were charged 13 cents a kilowatt hour for most hours except for 2:00 p.m. to 7:00 p.m. on weekdays, when the price went to 25 cents. On a few occasions the price was increased to 66 cents a kilowatt hour to mimic a period of special system needs. Evaluation indicated the program reduced peak demand by about 13 per cent.3

Results of the evaluation of 2003 programs is presented on www.energy.ca.gov.

Test results and results from general use of t-o-u might be quite different. Some customers might adjust their use to realise cost savings, while others might ignore the pricing changes. However, utilities, the Californian Energy Commission (CEC) and the Californian Public Utilities Commission (CPUC) are confident that, on the basis of the t-o-u pilots, this approach is effective.As a result, three major Californian utilities – PGE, Southern California Edison (SCE) and San Diego Gas and Electric (SDGE) are planning to replace conventional gas and electricity meters with up to 15 million t-o-u meters at a cost of around US$6 billion, beginning in 2006. The t-o-u meter expense will be offset to an unknown extent (depends on implementation policies and responses to them), by reduced peak usage: rate increases as a result of the meter rollout is expected by PGE to be small. In the period before t-o-u metering can make an impact, the California Energy Commission (CFC) estimates, that a 1 in 10 summer could result in a Southern Californian region shortfall of capacity of 2,000 MW (3.3 per cent) below demand by September 2005. Normal weather would not result in a shortfall and reserves would be adequate.As a response to the potential shortfall situation, the Californian Public Utilities Commission (CPUC) approved SCE’s request to implement additional energy efficiency programs aimed at reducing peak demand by 36 MW: insignificant compared to the potential shortfall. The decision orders SCE to expand four energy efficiency programs to immediately and significantly reduce peak demand – from residential customers and small, medium and large businesses.7

The programs:

  •  expand residential customers’ options for “instant rebates” – which are done at the point of sale – and are currently only available for compact fluorescent light purchases. The expanded program will include pool pumps and motors, refrigerators, air conditioners and whole house fans;
  • give  small  businesses  “no-cost”  lighting retrofits.  SCE    estimates    reaching approximately 10,000 customers through this effort; and
  •  allow larger business customers to apply for incentives of up to 100 per cent of the cost of the project on lighting retrofits.

Review of the Californian situation indicates that:

  • despite a range of in-place DSM/DR/DM programs the Californian system is still susceptible to disruptions;
  • t-o-u pricing may still some time off; and
  • the supply system is not being expanded at a sufficient rate to meet increasing demands.

 

The Californian Energy Commission (CEC) Integrated Energy Report8

This report, which is prepared every two years, with an update each alternative year, reports on the status of the State’s energy system and makes recommendations for action where it is deemed necessary.

Key issues identified in the 2004 Update are as follows:

  • implementation of the Energy Action Plan’s loading order strategy;
  • improved transmission planning is required to address inadequate transmission as it presents a significant barrier to accessing renewable energy sources critical to diversifying fuel sources;
  • reliability issues with ageing power plants;
  • the need for accelerated renewable energy developments; and
  • the need for acceleration of demand response programs that signal the actual price of electricity to customers in peak periods.

In the demand response area, the primary focus of this report, the 2004 Update calls for electrical utilities to aggressively implement the 2007 State-wide goal of reducing peak demand by 5 per cent. The 2004 Update appears to rely essentially on “dynamic pricing” (implemented through tariffs using t-o-u, interval meters) to meet this target.

Given the interval metering rollout schedule, likely rollout delays and uncertainty regarding peak tariffs and their impacts, it would seem that attention to other peak demand reduction and supply security are required if the target is to be attained. Thus, despite the 2000-01 disruptions actions to avoid a repeat the Californian system continues to be vulnerable to high (1 in 10) summer peaks.

The lesson for Victoria (and Australia generally) is that even after actual and significant supply disruptions, the implementation of preventive actions lags the requirements. Victoria/Australia has different circumstances: the private sector is responding on the supply side (but the Basslink delay reminds us of supply side reliance fallibility). BUT after five years of discussion, etc. little DSM to address peak loads has been implemented.9 (Would a serious disruption help?)

The 2004 Update, reviews progress on 2003 recommendations. In the DR/DSM/DM area:

(i)                        significant progress is reported on increasing energy efficiency funding and evaluation and monitoring of energy efficiency programs;

(ii)                       improved efforts are needed is reported on maximising energy efficiency of existing buildings; and

(iii)                     improvement is needed on rapid deployment of advanced (t-o-u, interval) meters and implementation of dynamic pricing tariffs.

In the case of (i) the Energy Commission recommended that the State10:

  • “Ramp up public funding for cost effective energy efficiency programs above current levels to achieve at least an additional 1,700 MW of peak electricity demand reduction and 6,000 gigawatts (GWh) of electricity savings by 2008.
  • Standardise and increase the evaluation and monitoring of energy efficiency programs to ensure that savings and benefits are being delivered. (Importance to be noted in VEES development.)

The State has made significant progress in this area, with the CPUC’s recent decision to adopt more aggressive goals for the investor owned utilities (IOUs) than the 2003 Energy Report recommended. These new goals, based on collaborative staff work between the Energy Commission and CPUC, require peak electricity demand reductions of 2,205 MW by 2008, exceeding the 2003 Energy Report goal by 505 MW, and energy consumption reductions of 10,489 GWh by 2008, exceeding the 2003 Energy Report goal by 4,489 GWh. These new goals will require approximately $522 million in annual funding by 2008 compared to the annual spending level of $348 million for 2004 and 2005.” And in the Executive Summary of the Update11 it is stated that “As recently as the 2000-01 electricity crisis, Californians embraced energy efficiency and demand response programs, reducing State demand by approximately 6,000 MW, more than 10 per cent of peak demand.”

In both cases (the 2003 Energy Report goals and the reductions to 2000-01) no evaluations are provided. This detracts from the credibility of the program results (see an outline of recent evaluation policies below). The Californian energy agencies (CPUC, etc.) proposed in a 2003 Energy Action Plan, in the energy conservation and resource efficiency area, that:

“California should decrease its per capita electricity use through increased energy conservation and efficiency measures. This would minimise the need for new generation, reduce emissions of toxic and criteria pollutants and greenhouse gases, avoid environmental concerns, improve energy reliability and contribute to price stability. Optimising conservation and resource efficiency will include the following specific actions:

  1. Implement a voluntary dynamic pricing system to reduce peak demand by as much as 1,500 to 2,000 megawatts by 2007.12
  2. Improve new and remodelled building efficiency by 5 per cent.13
  3. Improve air conditioner efficiency by 10 per cent above federally mandated standards.14
  4. Make every new state building a model of energy efficiency.
  5. Create customer incentives for aggressive energy demand reduction.
  6. Provide utilities with demand response and energy efficiency investment rewards comparable to the return on investment in new power and transmission projects.
  7. Increase local government conservation and energy efficiency programs.
  8. Incorporate, as appropriate per Public Resources Code section 25402, distributed generation or renewable technologies into energy efficiency standards for new building construction.
  9. Encourage companies that invest in energy conservation and resource efficiency to register with the State’s Climate Change Registry.”

The Decision builds upon Decision (D.) 04-09-060 and D.05-01-055 and an 21 April 2005 Decision 05-04-05, establishing the goals, policies and administrative framework to guide future energy efficiency programs funded by the ratepayers of the four largest investor-owned utilities (IOUs): Pacific Gas and Electric Company (PGE), San Diego Gas & Electric Company (SDGE), Southern California Edison Company (SCE) and Southern California Gas Company (SoCalGas).

D.04- 09- 060 established aggressive energy savings goals to reflect the critical importance of reducing energy use per capita in California. For the three electric IOUs, these goals reflected an expectation that energy efficiency efforts in their combined service territories should capture on the order of 70 per cent of the economic potential and 90 per cent of the maximum achievable potential for electric energy savings, based on the most recent studies of that potential. If successful, these efforts are projected to meet 55 to 59 per cent of the IOUs incremental electric energy needs between 2004 and 2013. On the natural gas side, adopted savings goals represent a 116 per cent increase in expected savings over the next decade, relative to the status quo. A three year cycle for updating savings goals, in concert with a three year program planning and funding cycle for energy efficiency (“program cycle”) was established and load reductions were included in savings goals.

In addition, an administrative structure for evaluation, verification and measurement (EM&V) was established to create a clear separation between “those who do” (the Program Administrators and program implementers) and “those who evaluate” the program or portfolio performance. (Victorian EES to note!) In particular, for program year (PY) 2006 and beyond, the Californian Energy Division will assume the management and contracting responsibilities for all EM&V studies that will be used to:

(i)                 measure and verify energy and peak load savings for individual programs, groups of programs and at the portfolio level;

(ii)                generate the data for savings estimates and cost effectiveness inputs;

(iii)              measure and evaluate the achievements of energy efficiency program, groups of programs and/or the portfolio in terms of the “performance basis” established under Commission-adopted EM&V protocols; and

(iv)               evaluate whether programs or portfolio goals are met.

The budget for EM&V was set, as a guideline, at 8 per cent of total energy efficiency program funds. (Note the significant resources that could be available for program evaluation at this level of funding.)15

Case study: Sempra Energy Inc/San Diego Gas and Electric (SDGE)16

Sempra/SDGE, serving a region in capacity constrained Southern California, operates a range of DSM programs, covering:

  • reduction of load during peak periods;
  • dynamic pricing; and
  • energy efficiency.

The utility claims over the past ten years to have cumulatively saved 1.9 million MWh, reduced peak load by 409 MW and provided cost savings to customers of over US$200 million.

2004-05 energy efficiency programs

Residential sector

Description of market segment:

Includes single family homes, condominiums, multi-family units, mobile homes and multi-family common areas.

The utility territory mainly has moderate coastal climate with high density housing and sparsely populated rural high desert and desert climates.

Provides electric service provision to approximately 1.2 million households.

Residential sub-segments:

  • single family customers;
  • multi-family customers; and
  • hard to reach.

Further segmented by end-use – air conditioners, all-electric homes.

Statewide residential rebates

Target market

All residential customers residing in SDGE’s service territory living in dwellings of 4 units or less, including condominiums and mobile homes.

 

Measures – rebates for:

  • Appliances;
  • Building shell – insulation;
  • Building shell – windows;
  • HVAC – air conditioning systems;
  • HVAC – controls;
  • HVAC – Ventilation systems;
  • Lighting – comprehensive products; and
  • Water heating – systems.

Industrial and commercial sectors

Commercial/industrial market segment includes over 138,000 electric meters and close to 30,000 gas meters.

Approximately 20 per cent of market consists of “large” customers – monthly kW demand above 500 kW.

Remaining 80 per cent of market consists of small and medium sized business with monthly demand of 500 kW or less.

  • Majority of the customer segment are considered “Hard-To-Reach”: rent or lease space; where English is the second language; businesses have less than ten employees; are outside urban San Diego, and annual electric demand is less than 20 kW or annual gas consumption is less than 10,000 therms, or both.
  • Almost 90 per cent of small and medium sized business customers have a monthly demand under 20 kW.

Industries are varied, including food service, property management, manufacturing, lodging, grocers and food growers.

Programs in these sectors include:

  • rebates for high efficiency HVAC systems and electric motor: delivered through system/product distributors;
  • provision of energy audits;
  • education and training programs for contractors, retailers, manufacturers;
  • building operator training and certification;
  • standard performance contract development and dissemination; and
  • incentives to participate in savings by design targeted at building owners and design teams to achieve “better than code” performance.

More information on the Sempra/SDGE program is set out in overheads from the utility’s Energy Efficiency Programs, Public Workshop, 3 March 2005. Although these programs are not targeted at peak load control, which will be addressed through t-o-u metering and tariffs, the SDGE’s comprehensive DSM measures that are summarised above:

  • can have a significant impact on peak loads; and
  • are well ahead of anything being implemented by Australian utilities.

It might be argued that the southern Californian situation has brought about such action and that program evaluation detail is lacking, but the SDGE programs (current and planned) indicate an innovative attach on energy efficiency improvement and peak load control that appears to be accepted by the government and its agencies.

Other USA state measures

A 2004 paper, Demand Response in the United States, prepared by the Wedgemere Group for the New Zealand Energy Efficiency and Conservation Authority (EECA) outlines DR/DSM/DM programs in a range of USA states17 and Ontario, Canada.

The outlines are a useful summary of these initiatives (websites are provided). TOU meters, coupled with dynamic pricing, is strongly supported in the EECA paper based mainly on the results of pilot programs in the USA: an average 0.3 demand elasticity is reported (for example, a 30 per cent demand reduction for a 100 per cent increase in price).

Program packages to address peak loads are not critiqued. Attachment B of the EECA report outlines reasons why new direct load control programs were not proposed in California.

The reasons provided are:

(i)                 the load impacts from these programs are already well understood;

(ii)               they limit customer choice: the utility determines the end-use (usually AC) and response level and does not allow customer overrides;

(iii)             they limit peak reduction potential to the chosen end-use load;

(iv)              they are inequitable because they offer a reward to owners of AC units, but not to non-owners; and

(v)                they are expensive because customers are paid even when the program is not used.

However, the possibilities for designing innovative load control programs in combination with t-o-u dynamic pricing and EEI programs are not considered in the EECA paper. This detracts from the usefulness of the paper from a policy perspective in the Victorian/Australian context.

Briefly, the reasons for rejecting direct load control are critiqued as follows.

(i)                 Load impacts from the earlier direct control may be well understood but are not for new designs of direct load control programs.

(ii)               More innovative designs can allow customer overrides: but if overridden full peak pricing would apply.

(iii)             They could be extended to other than A/C peak loads but A/C load is the load which is overwhelmingly weather dependent.

(iv)              They can be designed to reward non-A/C owners with lower rates than all A/C owners: that is, A/C owners taking direct load control would still pay more than non-A/C owners, but less than A/C owners not taking direct load control.

(v)                In combination with t-o-u meters, there is no reason why customers taking direct load control need not be paid when the program is not used.

Concluding comments

The United States experiences with measures for addressing peak loads are useful for analysis and consideration in the Australian/Victorian situation.

However, policy design here must be based on our particular circumstances.

Preliminary program designs for consideration and analysis (modelling, etc.) are set out in Attachment A.

Attachment A:

Scenarios for long run projections of Victorian peak demands

Introduction

This paper outlines potential measures for addressing summer peak load demands and suggests three scenarios for analysis of these measures.

For given weather patterns, population, income, economic trends, and consumer preferences, peak electricity demands will be driven by:

  • overall electricity prices (peak prices are considered separately), which rise to some extent as new plants are commissioned, but significant price increases will be mainly due to greenhouse (carbon price/permit) policies;
  • efficiencies of air conditioning units;
  • peak pricing policies; and
  • building trends.

Over the past five years, when it has been very evident that summer peak demands were increasing rapidly, the “non-policy” has been to build low capacity peaking plants or inter-connections. There has been virtually no policy directed at peak load control. This brief paper suggests how peak load control might be addressed. The study focuses on scenarios of policies to reduce (from BAU) peak demands in the residential sector: commercial and industrial sector analysis of peak demands requires separate analysis.

Three scenarios, two of which progressively reduce peak demands below BAU, are presented below for the 2005-50 period.

Under the BAU scenario electrical energy summer peak demand will continue to grow as population and incomes increase in each scenario. Income growth and consumer choice may translate into increases in average dwelling size, cooling of a greater proportion of space volume (whole house rather than one or two rooms), longer hours of operation and perhaps lower summer space temperatures. In the projections presented, these economic and social factors are held constant: further scenario development work would be required to assess their impacts.

Potential peak reducing policies

Emissions trading (carbon pricing)

Although the Federal Government continues to oppose the introduction of an emissions trading system for the pricing of carbon and trading of emission permits, States and Territories are continuing to work on the design of an ETS appropriate for Australia.

Action by the States/Territories and the possibility of a change in federal policy, suggests a carbon/permit price of $5/t CO2e by 2010 in a mild policy scenario and a price of $20/t CO2e by 2010 in a stringent policy scenario.

In the study these prices are assumed to remain over the 2010 to 2020 period, but increase to $10/t CO2e and $30/t CO2e respectively over 2020-2050 as the global greenhouse policy regime becomes more stringent, offset to some extent by technology advances which constrain the emissions permit price.

No explicit carbon pricing is included in the BAU scenario.

Peak load pricing and direct load control

In Victoria the installation of interval meters in all buildings will not be completed until about 2020. By 2013 only about one third of households will be fitted with interval meters (GWA, p.2918). Accordingly, unless there is a roll-out schedule change, universal time of use (TOU)/peak pricing in Victoria will not be possible until 2020.

There are several alternatives for direct control (for example through radio waves) of air conditioner loads and several trials are underway (New South Wales, South Australia, Western Australia) and a Ministerial Council on Energy (MCE) Committee is addressing the options. Work in this area commenced in 2000, but to date progress on developing policies and measures has been very slow.

Minimum energy performance standards (MEPS)

Levels for three phase air conditioning units were raised in 2004 and MEPS for single phase units introduced in 2004 are due to be raised in 2006 and 2007, with the final stage to match 2004 world’s best regulatory (not economic) practice in 2007. An indication of the impact of these MEPS changes is provided by GWA 2004, Table 4, p.23 and in the accompanying text.

There will be a rated performance improvement for the least efficiency split system units permitted to be marketed in Australia from April 2006. This improvement will be about 2 per cent for <4 kW and 5 per cent for >4 kW units compared with the average units sold in mid- 2004. Of current models available, about 10 per cent of <4 kW and 17 per cent of >4 kW would meet the proposed 2006 standard (GWA, 2004).

It is estimated (GWA, 2004) that the sales weighted efficiency for single phase units will then be 13-14 per cent higher, compared with 2004, than it would have been without the new 2006 MEPS.

World best practice for air conditioners is led by Taiwan and South Korea. The Australian MEPS lag the use of regulatory world best practice. As indicated above, MEPS applies, as the name implies, to the minimum acceptable rating (1- star) when the most efficient units (5-6 stars) are up to 40 per cent more efficient. The impact of higher air conditioning unit efficiencies on peak demands is debatable. Wilkenfeld, in a recent paper (GWA, 2004) claims, “where operation is intermittent and/or limited to one space, it is more likely that an increase in efficiency will lead to somewhat cooler internal conditions but have little effect on peak load”. (p.4, GWA 2004)

Why cooler internal conditions would result is not explained. In any case, this type of limited, intermittent situation is likely to become less important over time. MEPS levels could be raised by 2008 (or at least by 2010 -12) and/or greater efforts made to promote higher efficiency (5-star and higher) units.

Building trends: stock, sizes, retrofits and standards

The energy efficiency of buildings is increasing due to increased awareness of the net economic and environmental benefits achievable by improving the thermal efficiency of building envelopes and systems. Stock increases form a standard part of NIEIR’s projection methodology, but judgments on thermal efficiency trends must be made on the basis of policy and underlying trends.
In the case of new buildings, the Building Code of Australia (BCA) is moving to higher levels of thermal efficiency. From the early 1990s to 2004 there was only a slow and moderate increase in the thermal efficiency of new buildings. For example, in Victoria the 1992 thermal efficiency standard for new residences of about a 2 star rating had only increased to an average of about 2.7 by 2003. However, in 2004 a 4 star rating was mandated and on 1 July 2005 a 5 star rating will become mandatory. And work is being undertaken on a 6 star rating which is being achieved in a small proportion of homes.

Similarly, in the commercial sector movement to a 5 star rating for new buildings is likely (but not certain) in 2006. In the existing buildings area, retrofits are achieving higher thermal efficiency but the trend has not been quantified. Offsetting these trends, which reduce peak demands for a given stock, is an increase in building size (new or through refurbishment). Again, this trend has not been quantified.

No peak load pricing or direct load control is assumed in the BAU scenario. Faced with this uncertainty the following scenarios are suggested by NIEIR.

Scenarios for analysis of summer peak demands

Business-as-usual (BAU)

Over the past eight years, peak electricity demands have been increasing at about 4.0 per cent per year and are projected to increase at 2.6 per cent per year based on a 10 per cent POE through to 2015.

Although interval metering continues to be rolled out throughout the NEM, differential peak electricity pricing and specific load control measures are not introduced in this scenario. MEPS are held at 2004 levels in this scenario.

A 5 star requirement for new residences over the entire projection period (30 per cent net reduction in space cooling requirement compared with pre- 2005 new residences). No net increase in building size. No explicit carbon price is assumed in the BAU scenario.

Mild policy intervention

In this scenario the following new policy measures are introduced.

  1. An emissions trading system (ETS) is introduced in 2010 which results in a permit price of $5/t CO2e over 2010-2020 and an average electricity price increase of $6.5/MWh in Victoria over 2010-2020 (GHC4E), compared with 2005 levels. Over 2020-50 as the permit prices increase to $10/t CO2, average electricity prices increase by $10/MWh ($10/t CO2) as Victoria’s electricity greenhouse gas intensity reduces to an average of 1.0t CO2/MWh compared with 1.3t CO2/MWh over 2005-20).
  2. Air conditioner MEPS are accelerated resulting in an average 15 per cent increase in efficiency of new air conditioner units sold from 2008. (This means in effect, for example, that a previously rated 2 MW unit becomes a 1.7 MW unit from 2008 to 2020 compared with 2004.) This can be modelled by reducing the 2008 on growth in temperature dependent demands by 15 per cent.

By 2050 efficiencies are assumed to improve by 35 per cent (compared to 2004 levels).

(i)                 Peak pricing policies increase summer (October-April) peak prices by 30 per cent over 2005-20.

Customers are offered a lower price increase of 10 per cent if they agree to direct load control achieved through fitting devices to AC units which enable central control of AC units (for example through radio waves). Thirty per cent of customers accept this offer by 2020. Over 2020-50 peak prices increase by 50 per cent and customers are offered a lower price increase of 20 per cent if they agree to direct load control: 50 per cent of customers accept this offer by 2050.

(ii)               A 5 star requirement (30 per cent net reduction) for new residences from 2005-20 and 6-stars (40 per cent net reduction) from 2020 to 2050. No net increase in building size.

Stringent policy intervention

In this scenario the following policy measures are introduced.

  1. An ETS in 2010 results in a permit price of $20/t CO2e and an average electricity price increase in Victoria of $26/MWh over 2010-2020. Over 2020-2050 the average price increase is $40/MWh from a permit price of $40/t CO2e.
  2. Air conditioner MEPS are accelerated resulting in a 30 per cent increase of new air conditioner units sold from 2008 to 2020. (This means in effect that a previously rated 2 MW unit becomes a 1.4 MW unit.)

Over 2020-2050 average efficiencies of new air conditioner units increase by 50 per cent compared with 2004 levels.

  • Peak pricing policies increase summer peak prices by 50 per cent over 2005-2020.

Customers are offered a lower price increase of 20 per cent if they agree to direct load control as in 2. above. Fifty per cent of customers accept this offer.

Over 2020-2050 peak prices increase by 80 per cent and customers are offered a lower price increase of 30 per cent if they agree to direct load control: 75 per cent accept this offer.

5 stars for new residences over 2005-10, 6 stars over 2010-2020 and 7 stars (50 per cent net reduction from 2004 new residences) over 2020-30. No net increase in building size.

Note that in the latter two scenarios customer behavioural attitudes (for example in temperature control) to air conditioning is assumed to be similar to those in the BAU scenario. Behavioural changes scenarios could be introduced into the analysis but would require considerably more resources than proposed above.

Demand side management in California: current and proposed measures

Footnotes:

1        See Armstrong, G., California South: Coming to a Network Near You?, National Economic Review, No. 50, February 2002, for a review of the electricity situation which spawned many of these measures. 

2        Bob Thorkelson, Executive Director, Californian Energy Commission (CEC), Statement to Californian Senate Energy Utilities and Communications Committee, April 2005. 

3        Wall Street Journal, Rebecca Smith, 11 May 2005. 

4        The CPUC regulates the older so-called investor-owned-utilities (IOUs). Newer utilities are referred to as private utilities. In addition, there are municipally-owned utilities. 

5        Joint press release, 11 May 2005. 

6        CPUC press release, 5 May 2005 (www.cpuc.ca.gov). 

7        Thorkelson, op. cit. 

8        Californian Energy Commission, Integrated Energy Report, November 2004 update. 

9        Energy Australia time-of-use meter implementation 

Energy Australia announced in June 2005 that it will offer Sydney, Central Coast and Hunter Valley residents lower cost electricity in shoulder and off-peak prices via new “smart” power meters. The meters will allow Energy Australia to introduce different rates at different times. Lower prices will be offered in the morning and overnight, with customers able to reduce power bills by choosing the pricing period in which they use appliances such as dishwashers and air conditioners. 

The three tiered pricing structure will mean peak prices are charged between 2:00 and 8:00 p.m., “shoulder” prices from 7:00 a.m. to 2:00 p.m. and 8:00 p.m. to 10:00 p.m., and off-peak prices from 10:00 p.m. to 7:00 a.m. 

The new system will be phased in gradually, with new residential homes, those upgrading their electricity installation and big users with annual bills in excess of $4,000 the first to be offered the new meters. Existing customers can convert to the new meters if they pay for installation. According to Energy Australia, prices will be 70 per cent higher in the peak period than current prices, 20 per cent cheaper in the shoulder period and 60 per cent cheaper during off-peak times. 

The company claims a family with a $900 bill could save $100 by changing 5 per cent of their peak electricity usage to off-peak and another 5 per cent to the shoulder times. An audit of Energy Australia customers has found changing operating times for pool pumps, washing machines, dryers and dishwashers could have a marked impact on bills.

10      2004 Update, p.54.

11      Ibid, p. xiii. 

12      California continues to actively evaluate and implement such pricing systems under a CPUC rule-making (R.02-06-001) edict. 

13      The Energy Commission’s new building standards, to be adopted in 2006, when combined with training and enforcement, are expected to reduce energy needs in new buildings by approximately 5 per cent. 

14      New federal appliance standards will increase air conditioner efficiency by approximately 20 per cent by 2007. However, if California were granted a waiver from federal standards, by 2007 the CEC estimates that California air conditioner efficiency could increase by another 10 per cent. 

15      Interim Opinion: Updated Policy Rules for Post-2005 Energy Efficiency and Threshold Issues Related to Evaluation, Measurement and Verification of Energy Efficiency Programs, Decision 05-04-051, 21 April 2005. 

16      Summary of Sempra Energy/SDGE presentation, Energy Efficiency Programs, Public Workshop, 3 March 2005. 

17      The paper regards Demand Response (DR) as only applying to peak load reduction measures, including distributed generation (DG), but including EEI in only a long term sense. This definition is not universally accepted. 

18      A National Demand Management Strategy for Small Air Conditioners, for the National Appliance and Equipment Energy Efficiency Committee (NAEEC) and the Australian Greenhouse Office (AGO), November 2004 (GWA 2004).

Economic Overview (NER 58)

National Economic Review No. 58 September 2005

The National Economic Review is published four times each year under the auspices of the Institute’s

Academic Board.

The Review contains articles on economic and social issues relevant to Australia. While the Institute endeavours to provide reliable forecasts and believes material published in the Review is accurate it will not be liable for any claim by any party acting on such information.

Editor: Dr A. Scott Lowson

National Institute of Economic and Industry Research

This journal is subject to copyright. Apart from such purposes as study, research, criticism or review as provided by the Copyright Act no part may be reproduced without the consent in writing of the Institute.

ISSN 0813-9474

Economic overview

Peter Brain, Executive Director, NIEIR

Abstract

Peter Brain assesses the Australian economy and describes alternative scenarios

Although the GDP growth for 2003-04 was 3.6 per cent, this represented a relatively poor performance.

The GDP growth rate of 3.6 per cent for 2003-04 was the same as earlier projections. However, it represented a relatively poor performance. The reason for this assessment is due to the fact that over 2003-04 the Australian farm sector recovered from the drought. Farm product in 2003-04 grew by 27 per cent, adding 0.7 per cent to GDP growth. However, non-farm GDP grew by 3 per cent for 2003-04 despite a 5.6 per cent private consumption growth which represents the highest rate of growth for a number of years. Moreover, the growth rate of all the private investment components was 6 per cent or greater.

The reason for the relatively poor GDP growth outcome is, firstly, the poor performance of exports and, secondly, the growth in imports. There is a lag between farm production recovery and exports so the growth in exports resulting from the farm recovery will occur in 2004-05.

In 2003-04 imports grew by 13.1 per cent, only slightly below the growth in 2002-03. This represents a growth in import penetration across a wide range of sectors, including clothing, textiles, motor vehicles, chemicals and machinery. Imports represent one quarter of GDP. Hence, a 13.1 per cent import growth rate means that the growth in imports over 2003-04 reduce GDP by 2.5 per cent from what would otherwise have been the case if imports had growth in line with GDP.

Over the last two years in particular, the growth in imports has been a major negative factor in determining

Australia’s growth performance.

Australia’s exports performance has also been poor but will recover over the next three years.

In the few years since 1999-00, the value of Australia’s non-resource based exports has been flat. That is, no change has occurred. This is despite the value of trade in the Asia-Pacific region for non-resource based products growing between 30 and 40 per cent over the past four years.

In 2004-05 exports of goods and services are expected to grow by 5.1 per cent, in part due to the recovery of the farm sector. Exports will also recover over the next two to three years because of the coming on-line of major resource projects that were commenced in 2002 or 2003. The most important of these will be the fourth liquefied natural gas (LNG) train on the North West Shelf. In 2006 the Darwin LNG train will come on-line.

Both the United States and Australian dollars will devalue over the next five years relative to our trading partners.

Exports may well recover, but without a substantial devaluation of the Australian dollar, import growth will continue to outstrip the growth of exports. With the upswing in the world interest rate cycle now occurring, the continuation of the current growth in imports would lead to an Australian current account deficit of around 7 per cent of GDP. To hold the current account deficit at the 5 per cent level, which is the projection to 2008-09, it is necessary for the Australian dollar to devalue,

in weighted average terms of around 15 per cent over the 2006 to 2009 period. This is built into the projection.

It can be seen from Table 1 that the United States/Australian exchange rate stays relatively unchanged over the projection period. The projection also allows for the outcome that the United States dollar devalues 20 per cent against the Euro, yen and yuan over the projection period. Because Australia maintains parity with the United States dollar, it follows that there is an equivalent devaluation of the Australian dollar against these currencies. The appreciation of the yuan against the United States dollar is also assumed to trigger the appreciation of other Asian currencies against the United States dollar.

It is the devaluation of the Australian dollar that leads to a more subdued growth rate for imports over 2008 and 2009.

The recent evidence is that the downside phase of the dwelling cycle has commenced.

It has long been NIEIR’s contention that the down-phase of the current dwelling cycle would only commence when significant growth in established house prices ceased. By the June quarter 2004, established house prices had stabilised with a fall in established house prices in Sydney offset by more moderate price growth elsewhere. Moreover, the trend in approvals and the financing of dwellings for new construction all point to falls in dwelling construction over the next two years. Over the next two years the cumulative decline in housing construction is projected to be 18 per cent.

The borrow and spend behaviour of households is now reaching its peak. Household balance sheet constraints will be a negative factor for growth for the foreseeable future.

The ending of the established house price boom will also lead to a curtailment of a key driver of recent Australian economic growth, namely household borrowing to support consumption expenditure.

The growth in established house prices since 1996 resulted in the ratio of household net worth (the value of the housing stock plus financial assets less financial liabilities) increasing from 6 to 7.8 by June 2005 (Figure 2). From Figure 4, this allowed households to borrow to fund a borrowing gap which has reached 15 per cent of disposable income by June quarter 2004. The borrowing gap represents the difference between consumption expenditure and discretionary income. Discretionary income is significantly smaller than household income in the national accounts because it includes superannuation contributions and superannuation interest, which represents income that is not available for current consumption.

From Figure 3, by the June quarter 2004 the build up in debt to fund the borrowing gap (as well as the high level of housing investment) drove the household debt to net disposable income ratio to 163 per cent. In the June quarter 2002 the rate stood at 137 per cent.

From Figure 1, the household debt service ratio now stands at 25 per cent of disposable income, the highest on the historical record.

The combined impact of stable (or falling) house prices, high debt service and debt-income ratios will, at the most optimistic, force households to hold the borrowing gap at around 15 per cent of income. This will force consumption expenditure to grow in line with household disposable income, which in turn will reduce the rate of growth of private consumption expenditure to between 2 and 3 per cent over the medium term.

Even with modest consumption growth, the debt-income/debt-service ratio will continue to rise. A recession is likely at some point before 2010.

If the borrowing gap is held at 15 per cent, the debt-income ratio will still increase by around 7 percentage points per year. By 2009, given the projection in Table 1, the debt to income ratio will reach 200 per cent. If households decide to stabilise their debt-income ratio then the household savings ratio will have to rise to 6 to 8 per cent. Household consumption would most likely fall and the economy would experience a recession, probably a severe recession.

However, given the forecast methodology outlined above, this aspect has been translated into a lower trend rate of growth rather than a recession and this aspect makes the low case projection of more interest than the high case projection.

Fiscal stimulus will support the household sector in the short term.

The position in the short term is not as bleak as the borrowing gap would suggest because of the strong fiscal stimulus being given to the economy. The May 2004 Federal Budget and the election promises of October 2004 will give a stimulus of around 1 per cent per annum to household income over the next two to three years. This will probably be enough to partially offset the constraints of the household debt-service ratio. Beyond 2007, if a severe recession is to be avoided, further significant fiscal stimulus will be required. That is, as the growth in household debt slows, public sector new borrowings will have to increase significantly.

The alternative scenarios

The problem for Australia is that Australia is not the only economy with households with large amounts of illusionary wealth created by housing price bubbles. The same is true in North America, the United Kingdom and some Western European economies. An economy that is an indicator, in terms of a low scenario over the medium term, is the Netherlands. The Netherlands was a fast growing economy over the second half of the 1990s, in part driven by rapid increases in borrowings funding a house prices-wealth creation consumption boom. In 2001, house prices stabilised due to tightening monetary policy. In 2003 the economy was in recession with private consumption falling by 1.5 per cent, the largest fall since World War II.

For the Netherlands the catalyst was tightening European monetary policy over 2000. For Australia the likely trigger for a low scenario is also most likely to be an external shock such as illustrated in Table 2. There are a number of potential shocks with good probabilities of occurring over the next two to five years. They are listed in the Table.

The reason why a transition path from the base to low scenario is likely to be associated with an external catalyst is that there are two factors that would allow policy authorities to keep the economy on the base scenario trajectory despite increasing constraints in growth. These are:

  • strong public sector balance sheets which would allow fiscal policy to be expansionary for a decade or more; and
  • the potential for Australian nominal interest rates to be lowered by between 1 and 2 percentage points.

This cushion would allow the base scenario to be achieved if the world economy remained supportive.

Unfortunately, because of vulnerable households in a number of major economies, any negative shock to the world economy is likely to trigger the ushering in of a long period of low growth for Australia, in particular, and many parts of the developed world in general. In short, the low scenario, at least to 2012 or thereabouts, does not have a low probability of outcome.

The high scenario assumes the most optimistic outcomes for the world political economy.

Australian energy trade, 2004-10

ABARE and NIEIR analysis and estimates of Australian energy trade trends are presented below. Over the period there continues to be an energy trade surplus with projected increases in net oil imports being more than offset by coal, natural gas and uranium export increases.

In 2004-05 the trade surplus, at a projected $7.4 billion (NIEIR/ABARE), will be about $2 billion higher than in 2003-04 due to higher thermal coal exports (tonnes, prices) and higher LNG exports.

Table 1        Major economic aggregates: financial year averages (annual per cent rate of change)

1998-99

1999-00

2000-01

2001-02

2002-03

2003-04

2004-05

2005-06

2006-07

2007-08

2008-09

International
G7 real GDP

2.3

3.4

2.4

0.2

1.9

3.1

3.3

2.6

2.0

2.5

2.2

Trade partners real GDP

1.4

5.6

3.7

1.6

3.6

4.9

5.0

4.2

3.7

4.4

4.3

G7 CPI

1.0

1.5

2.0

1.1

1.5

1.5

1.8

1.8

1.7

1.9

2.0

Trade partners CPI

6.0

2.1

2.7

3.1

2.9

2.8

2.9

3.0

3.0

3.3

3.3

GDP and components
Private consumption expenditure

4.8

4.1

2.9

3.3

3.8

5.6

4.1

2.8

2.8

2.6

3.5

Non-dwelling construction

13.2

-8.6

-18.4

12.3

32.0

12.0

5.8

5.4

2.1

-0.2

6.3

Equipment

1.0

11.1

6.1

6.1

16.7

6.0

9.6

5.4

5.3

5.5

5.3

Housing

7.6

14.4

-20.8

19.2

15.4

7.7

-6.3

-10.5

4.3

8.5

6.9

Public consumption expenditure

4.0

2.9

2.0

2.1

4.4

3.3

4.2

3.5

4.2

4.5

3.6

Public investment expenditure

3.9

7.6

-11.9

0.9

6.4

5.3

6.1

3.3

3.7

4.1

-0.9

Stocks and other (% points)

-1.3

0.6

-0.2

-0.4

0.1

-0.7

1.0

-0.1

-0.3

-0.2

0.4

Exports

2.0

9.6

7.3

-1.1

-0.5

0.9

5.1

6.6

4.8

4.3

2.3

Imports

4.8

12.9

-1.3

2.2

13.5

13.1

5.1

3.7

4.1

4.3

4.4

GDP

5.3

3.8

2.0

3.9

3.1

3.6

3.0

2.9

3.1

3.1

3.5

Farm GDP

13.7

9.9

-0.8

4.6

-25.2

26.8

-1.9

-2.3

4.9

5.4

0.0

Non-farm GDP

5.0

3.6

2.1

3.9

4.1

3.0

3.2

3.0

3.1

3.0

3.5

Dwelling sector
Commencements

-34.1

45.3

3.4

-0.8

-13.3

-7.9

16.3

-5.1

0.4

Labour market
Employment

2.0

2.1

2.1

1.2

2.5

1.8

1.9

1.8

1.8

1.7

2.4

Unemployment rate

7.4

6.6

6.4

6.8

6.2

5.8

5.6

5.6

5.6

5.7

5.7

Population

1.1

1.2

1.4

1.2

1.2

1.3

1.2

1.2

1.2

1.2

1.1

Wages and prices
Average weekly earnings

3.5

2.9

4.8

5.3

5.3

5.6

4.2

3.9

4.2

4.7

4.8

CPI

1.3

2.4

6.0

2.9

3.1

2.4

2.6

2.5

3.2

3.3

3.1

Real household disposable income

4.4

3.9

4.6

1.9

1.3

5.1

3.6

2.4

1.8

2.3

3.5

Finance
90 day bill rate (%)

4.9

5.6

5.8

4.6

4.8

5.3

5.4

5.1

5.0

5.1

5.5

10-year bond yields

5.4

6.5

5.8

5.9

5.3

5.7

5.8

5.6

5.8

5.9

6.1

$US/$A

0.63

0.63

0.54

0.52

0.58

0.71

0.73

0.74

0.72

0.70

0.71

External sector
Current account balance ($billion)

-33.6

-32.6

-18.0

-20.6

-40.3

-47.4

-45.7

-46.2

-48.2

-50.2

-55.5

Current account balance (% of
GDP)

-5.5

-5.1

-2.6

-2.8

-5.2

-5.7

-5.2

-5.0

-4.9

-4.8

-4.9

Table 2 The alternative scenarios: potential drivers

Low scenario

  • 1. Terrorist strike against oil supply infrastructure in the Middle East. Oil price spikes to US$70 and above. World growth falls to zero 2005-2007.
  • 2. Sharp devaluation of US dollar. Capital flight. US dollar devalues 40 per cent. Inflationary pressures in United States forces interest rates up to 8 per cent. Severe United States recession followed by a decade of slow growth.
  • 3. Slow growth in developed world results in protectionist measures against China/India. Environmental and economic problems, drying up of capital inflow from low Chinese growth after Beijing Olympics. China becomes isolated and military tensions in North Asia return to 1960 levels.
  • 4. The rise in world interest rates and slower world growth results in Australia’s current account deficit rising to 8 per cent of GDP. Australian interest rates are raised to reduce domestic demand. Consumption expenditure falls 5 per cent over two years. Capital flight from Australia devalues currency by 40 percent in trade weighted terms. Inflation of prices critical in demand expansion measures for some time. Governments are slow to adopt expansionary fiscal policies.

High scenario

  • 1. Iraq stabilises and political solutions reached in Middle East between Jews and Israelis.
  • 2. Terrorism reduced to minimum levels. Oil price returns to US$15 to US$25 range.
  • 3. Euro Asia adopts expansionary monetary policies. Current high savings ratios allows the borrow and spend dynamic to drive above average European growth rates for 10 to 15 years. European growth takes pressure off United States balance of payments. Reformist United States government from 2008 raises taxes and stabilises the fiscal situation. China continues to open up and changes from a commercial economy to a market economy.
  • 4. Political transition to a basic democratic framework. World trade expands at 6 to 9 percent a year resulting in a rapid convergence of living standards between India/China and the developed economies.

Table 3        Australian energy trade trends

(A$ billion nominal)

Commodity

2004-05

2010

Coal (thermal)1

6.3

7.5

Natural gas (LNG)2

3.5

5.1

Uranium3

0.5

0.6

Oil (crude and products)4

-2.9

-5.0

Total

7.4

8.2

Notes:

  1. 1.2004 average price = $50/t.
  2. 2004 exports = 8 Mt; 2010 exports = 20 Mt.
  3. Contract prices (A$/lb U3O8) rising by 20 per cent, 2003-10.
  4. 2004; 80 per cent net self sufficiency at A$60/b.
    2010; 65 per cent net self sufficiency at A$50/b.

no58-graph

no58-graph-2

no58-graph-3

no58-graph-4