Electricity Forecast: Disruptive and More Centralized
The escalating controversy over recent bankruptcies at solar start-up companies like Solyndra and Abound Solar is a distraction from real story about the solar power industry. It is not only here to stay, but poised for rapid and relentless growth over the next 20 years. Make no mistake about it: the future of widespread solar poweris now inevitable. The same can be said about electric vehicles, energy storage and fuel cells, but perhaps the most underappreciated change coming down the pike is the advent of market-aware energy consuming assets. Here are five reasons why these changes are inevitable:
Human intelligence - and the ability to build upon past discoveries – is increasing every day. It’s the positive side of Toffler’s Future Shock. With computational capabilities, we can run virtual tests, in parallel, on many different technologies, and learn far more quickly than ever before. And we can instantaneously disseminate knowledge. Thus, technological advancement that might have taken decades to achieve in the pre-computer age may now take only a few years. For proof, look at what is happening in solar with all of the different technologies. Solyndra is an example of this, and a good one. While pointing out the weakness of investing too heavily in a single technology, the case also highlights a reason for optimism: Solyndra lost because (forget the Chinese subsidies for a minute) the competing technologies advanced too quickly and became too cheap too fast for Solyndra to keep up. Other technologies are similarly advancing, though perhaps not as quickly.
Our grid is aging. Much of it is older than I am. And it is extremely expensive to replace. Those high avoided costs create an opportunity for efficiency,decentralized supply solutions, and responsive demand assets.
Prices. We have seen a bottoming out of electric energy prices. Natural gas is unlikely to ever get cheaper than it was last year when we were punching so many holes into the shales that short-term supply overwhelmed demand. Many people seem to confuse short- and long-term elasticities. Over time, new demands for shale gas will emerge; gas-powered trucks and vehicles, LNG exports (Cheniere is the first of many in line in the permitting process), and new gas-fired power plants. The EIA projects retirement of 49 gigawatts of electric demand through 2020. Gas-fired generators will meet a large portion of that demand..
Carbon. We have fallen into a believer/non-believer dialogue (if one can call it that), which appears more driven by ideology than science. Yet, assuming the majority of scientists are correct, the dynamic will not go away, ideology notwithstanding. This July was hotter in the US than the year of dust bowl; one month doesn’t make a climatic trend. But ten years just may. The odds appear to favor hotter, drier, and more volatile weather. Sooner or later (if you agree with the Economist), that leads to higher energy prices through some sort of carbon based pricing.
Economic growth. Even an anemic 2% economic growth rate would see a doubling of electricity consumption in 35 years. 3% moves it to 23 years. The average annual rate from 1948 to 2009 was 3.28%. A doubling of consumption and associated infrastructure and fuel costs suggests higher underlying fuel AND infrastructural costs.
So the stage is set for potentially higher prices (absent some black swan event we cannot foresee), and improved relative economics of new distributed technologies whose costs continue to come down. We can expect to see deployment of these technologies in four distinct areas: 1) where subsidies and regulatory decisions provide early safe havens; 2) where high avoided costs improve the economics (the island of Omotepe in Lake Nicaragua is said to have 60 cents/kWh electricity rates – the race for solar should be on!); 3) where volatility is most pronounced; and 4) where distributed technologies can add the highest value in terms of what and when avoided resources are displaced.
The key to optimizing many of these investments, irrespective of technology, is to tie them into an economic context. In other words, we need to make them market aware, and profitable. In so doing, we can optimize the size of the assets during the initial investment phase, and their performance on a daily basis. Let’s take gas-fired back-up generation in Texas for example. Assuming a consumer requires 1 MW of back-up gen for a business-related need such as reliability, the question during the initial assessment phase should focus around trade-offs between size of the unit and value of participation in energy markets. Given market volatilities, capital costs to supersize the unit, and underlying fuel costs, the optimal investment could be a unit x% larger, which could sell into the market whenever the real-time price exceeds a specific price. Such decisions require analytical capabilities, an understanding of past and expected future market dynamics, and an awareness of the regulatory environment. Most of all, it requires knowledge of real-time market characteristics.
The future grid will be characterized by market awareness of many decentralized resources, and require an intelligence which does not currently exist.