Friday, April 10, 2015

Customers Will Leave the Grid Behind in Droves Over Next 10-15 Years

Rocky Mountain Institute has just published a fascinating study on the future of "grid load defection" in the U.S.A. under likely pricing scenarios.  You can download it here:

http://blog.rmi.org/blog_2015_04_07_report_release_the_economics_of_load_defection

There is now significant concern in the U.S. -- and there SHOULD be significant concern in Japan -- about the likely failure of the traditional utility business model, as revenues decline once customers can self-generate electricity on an economic basis using solar PV and, eventually solar PV + battery storage.

One response is to charge customers a fixed amount for maintaining the grid, even if they reduce their consumption significantly based upon self-generation.  RMI's study suggests that such approaches only prolong the inevitable.  Over the next 10-15 years, almost everywhere in the U.S. a combination of solar PV + battery storage will become economically optimal, causing customers to purchase much less power and pushing down utility revenues.

The RMI study posits maximum potential customer defection in the Northeast U.S. at 50% of residential and 60% of commercial customers by 2030.

They highlight the need for new utility business models and new regulatory approaches to avoid this.  Of course, the faster on-grid electricity prices rise ... the faster defection will occur.  And the faster solar PV and storage costs decline ... well, you get the picture.

According to RMI, "although they could represent significant load loss, customers’ grid-connected solar-plus-battery systems can potentially provide benefits, services, and values back to the grid, especially if those value flows are monetized with new rate structures, business models, and regulatory frameworks."

But there is a major risk of a huge new group of centralized generation "stranded assets".

----------------

Now think about the situation in Japan where on grid power is much more expensive, and the cost is going up much more rapidly, and the current central planning process to decide on "energy mix" is giving a major role for new coal-fired generation and nuclear -- two generation sources that are not currently "economic" and so are just not being built in the U.S. ... and in many cases being mothballed.  What is the likely result?

Renewables in ... Texas

Japanese involved in the energy business and energy policy with whom I speak often marvel at the U.S. "shale gas revolution".  They envy the U.S. access to cheap fossil fuels.  Yes, natural gas is dirt cheap and the U.S. is now, again, the #1 oil producer in the world.

But they are much less aware that the fastest growing area of electric power generation in the U.S. is not gas, but renewables -- solar and wind.  Well of course, they think, in someplace like California where liberals from Hollywood support a governor like Jerry Brown and drive aggressive renewables targets.

But wait, renewables in Texas -- the heart of the "oil patch"?

Indeed, Texas is the #1 producer of wind power in the U.S.A.

And this article in Scientific American highlights how solar PV is going to play a major role in Texas.

Austin, Texas plans almost 1GW of utility scale solar supply by 2025.  And they plan to do it while maintaining affordability -- requiring lower than average utility bills and holding annual increases below 2% (the Kuroda BOJ inflation target, coincidentally).

Well, the article focuses on liberal Austin, Texas, home to University of Texas, Austin and the State Capital.  But Austin is a city of almost one million population, and growing very rapidly.  Austin's electric utility is at 25% renewable energy supply today and plans to be at 55% renewables by 2025.

It is indeed a Brave New World when a major city in the heart of the oil patch plans to be a 55% renewable electricity supply within ten years.

Meanwhile, Japan is building lots more coal plants and hopes to get to ... maybe 25% renewables by 2030?  And they divide the world into "baseload" and "non-baseload"?  Huh?