Understanding impact of cap and trade (Waxman-Markey) on IT departments and networks

[For further in depth analysis on this subject please see the upcoming Educause Review special publication on this topic and presentations by Dr. Larry Smarr and yours truly at Educause summit in Denver in November.

There has been a lot of discussion about climate change and what IT departments should do to reduce energy consumption. Most of this is being driven by corporate social responsibility. But a few organizations are undertaking processes to understand the impact of cap and trade on the bottom line of their IT and network operations. When the real cost of cap and trade starts to be felt a lot of organizations will be looking at their IT departments as the low hanging fruit in terms of reducing energy consumption and concomitant GHG emissions.

Only marginal energy reductions are possible with traditional electrical hogging sources such as lightning, heating, air conditioning etc. IT holds out the promise of much more significant savings because of its inherent flexibility and intelligence to support "smart" solutions. Several studies indicate that ICT represents at least 30% of the energy consumption in most organizations and it is estimated as much as 50% within certain sectors such as telecoms, IT companies themselves and research universities. Hard, quantifiable data is difficult to find - but CANARIE is funding 3 research projects to do a more detailed analysis of actual electrical consumption by ICT and cyber-infrastructure for at least one sector in our society - research universities. (Preliminary results are already pretty scary!)

To date the various cap and trade systems have had little impact because either emission permits have been effectively given away, or the underlying price of carbon has had a negligible impact on the cost of electricity. This is all about to change. First with Waxman-Markey bill (HR 2454) now before the senate and the move to auction permits in the European Trading System (ETS). Even if the Waxman-Markey bill fails to pass in the Senate, there are several regional cap and trade initiatives that will be implemented by US states and Canadian provinces in the absence of federal leadership. So, no matter which way you cut it, electrical costs for IT equipment and networks are projected to jump dramatically in the next few years because of cap and trade. On top of that there may be energy shortages as utilities move to shut down old coal plants where it does not make economic sense to install carbon capture sequestration (CCS) systems to comply with the requirements of these cap and trade systems.

The US Environmental Protection Agency (EPA) has done some extensive modeling and economic analysis of the impact of the Waxman-Markey bill. It is probably the best source for a general understanding of how various cap and trade systems around the world are going to affect IT operations. Even though some of the particulars of the bill may change in the US Senate, the broad outline of this bill as well as those of other cap and trade systems will remain essentially the same. Details of the EPA analysis can be found here:

http://www.epa.gov/climatechange/economics/economicanalyses.html

Surprisingly there has been little analysis by the IT industry sector itself on the impact of cap and trade on this industry. IT may be the most significantly affected because of its rapid growth and its overwhelming dependency in several key sectors of society such as university research, banking, hospitals, education, etc. Although IT overall only consumes 5-8% of all electricity depending on which study you use and contributes 2-3% of global CO2 emissions, IT electrical consumption is over 30% in most businesses and even greater amounts at research universities. What is of particular concern is that IT electrical consumption is doubling every 4-6 years and the next generation broadband Internet alone could consume 5% of the world’s electricity. Data centers as well are project to consume upwards of 12% of the electricity in the US.

There are number of important highlights in the Waxman-Markey bill that will be of significance to IT departments and networks:

1. The proposed cap reduces GHG emissions to 17% below 2005 levels by 2020 and 83% by 2050.

2. Most of the GHG reduction will be from the electricity sector and purchase of international offsets in almost equal portions.

3. GHG emissions from the electricity sector represent the largest source of domestic reductions - although transportation accounts for 28% of emissions in the US, only about 5% of the proposed reductions will come from that sector and expected to raise gasoline prices by only a paltry $.13 in 2015, $.25 in 2030 and $.69 in 2050 (Much to the relief of the oil industry, Canada’s tar sands and owners of SUVs)

4. The share of low or zero carbon primary energy rises substantially to 18% of primary energy in 2020, 26% by 2030 and 38% by 2050, although this is premised on a significant increase of nuclear power and CCS. True renewables only make up to 8% in 2015, 12% in 2020, and 20% in 2030

5. Increased energy efficiency and reduced energy demand simultaneously reduces primary energy needs by 7% in 2020, 10% in 2030, and 12% in 2050.

As you can imagine there are many uncertainties and controversial assumptions that affect the economic impacts of H.R. 2454 and many other cap and trade bills. Briefly these are some of them:

(a) The degree to which new nuclear power and CCS is technically and politically feasible. HR 2454 assumes a dramatic increase in nuclear power and deployment of CCS. If either fails to materialize then the GHG reduction targets will not be met. Assumption of growth in nuclear power is particularly suspect as any new nuclear plants in the foreseeable future will be first needed to replace the many aging systems now at the end of their operating life.

(b) The availability of international offset projects. Given the controversy that already exists over international offsets many question the assumptions of being able to purchase this volume offsets particularly when every other country with a cap and trade system will be pursuing this same market.

(c) The amount of GHG emissions reductions achieved by the energy efficiency provisions. In the IT sector in particular growth of IT products and services may simply outweigh any gains made in efficiency.

Although the impact of HR 2424 on consumer electrical costs will be minimal, its a different story for business and industry users. The EPA estimates that the "average" price of electricity will increase by 66% for commercial users. But there will be huge regional variances in these prices depending upon the amount of electricity that is produced from coal without CCS. In those regions largely dependent on coal generated electricity the cost increase will be almost entirely dependent on the market price of carbon.

If your electricity is mostly generated by coal, which includes most of the mid-west in USA and western Canada, then a rough rule of thumb is 1000g of CO2 is produced for every kilo-watt hour of electricity which results nice easy one to one conversion of annual hourly consumption to metric tones of CO2. A typical research university has a 40 MW utilization which translates into about 350,000 MWhr of consumption. This would result in 350,000 mTCO2e. If carbon trades at $25/ton then the increased cost to the institution will be in excess of $8 million per year.

However if many of the assumptions in the Waxman-Markey fail to come to pass, particularly the availability of international offsets then cost of carbon could jump dramatically. (To protect against this the US senate is proposing a “collar” to limit variability in price of carbon). The EPA analysis has various projections for carbon, and depending on the scenario the cost could go up to as much as $350 per ton, if the objective of 17% in GHG reductions are going to be achieved by 2020 and 83% by 2050. The Nicholas Stern report in the UK suggests that carbon must trade at a $100 a ton to achieve meaningful GHG reductions.

One of the main concerns of the Waxman-Markey bill is that it is too little and too late. More and more evidence points to much more rapid warming of the planet than even the most pessimistic computer models have forecast. Although we had a wet and cool summer in eastern North America average global sea temperatures set a new high record this year. The latest study from UK Meteorological office, that incorporates CO2 feedback cycles for the first time, suggests that US could warm up by 13-18F and the Arctic by 27F by 2060. The bottom line is that Waxman-Markey is just a starting point to probably much more stringent GHG reduction policies. The IT sector needs to get prepared for this worst case eventuality. If nothing else it should be part of any disaster planning scenario. This will be the mother of all disaster planning scenarios as opposed to other natural disasters that might affect IT operations it will be long term, if not effectively permanent.

However there is some good news for the ICT sector. One of the requirements of the Waxman-Markey (Title I, Subtitle A, Sec. 101) requires retail electricity providers to meet a minimum share of sales with electricity savings and qualifying renewable generation funded through purchase of offsets or other credits. Nominal targets begin at 6% in 2012 and rise to 20% by 2020. The ICT sector is probably the best qualified to take advantage of these energy requirements by adopting follow the wind/follow the sun architectures and relocating, as much as possible, computers and databases to renewable energy sources. The key to take advantage of these opportunities is to start planning now. Several papers from MIT and Rutgers indicate that savings of up to 45% in electrical costs are possible with such a strategy. These savings will be more significant with the advent of cap and trade.



-- BSA]
--------------
Bill.St.Arnaud@gmail.com
www.canarie.ca/~bstarn
skype: pocketpro
blog:http://billstarnaud.blogspot.com/