Regulatory uncertainty and industry investment: a new approach
By Sookhan Ho, Pamplin College of Buiness
How does regulatory uncertainty affect power plant investment and construction?
The answer could have implications for the banking, auto, and other industries far beyond electricity generation and supply, says Carl Ullrich, assistant professor of finance.
In studying this issue, Ullrich is taking an innovative approach, using data reported about every power plant in the United States to test models developed for studying financial markets. His findings, he hopes, will shed light on the broader impact of regulatory uncertainty.
In corporate finance academic circles, Ullrich says, the idea that uncertainty delays investment has been thoroughly discussed. Many theoretical studies have been done on “real options,” which are assets, such as machinery, factories, or power plants, that can be treated like financial options (securities that give buyers the right, but not the obligation, to buy or sell an asset at a set price at a later date).
“But there’s been very little empirical work to verify or test these real options models of how uncertainty affects investment, mostly because we haven’t had a lot of good data.” Until now.
Treasure trove of data
The data Ullrich found is in fact publicly available from the Energy Information Administration (EIA) website, but “no one has ever used it in this way before.” And it’s quite a treasure trove of statistics — almost 20 years worth of facts and figures on all U.S. power plants, not just the ones that were built, but also the ones that were proposed and the ones that were expanded or otherwise modified. It’s all information that plant owners have to submit every year to the federal government.
This data makes the research project exciting and unusual, says Ullrich, who is working with finance Associate Professor Randy Billingsley and doctoral student Jaideep Chowdhury on different studies that examine the impact of regulation on overall changes in planned power plant investment and on start-up and shut-down decisions.
“We have the kind of information most finance data sets do not have — in particular, data related to planned changes, things that don’t exist yet,” he says. “It takes a long time to build a power plant, so a regulatory decision today — we might not see that reflected in what’s actually built for a long time. But we can see from the data how plans have changed in reaction to regulation.”
Regulatory uncertainty is an important issue, he says. “I’m not talking about a high level of regulation versus a low level, but about what happens if regulators dawdle, if they leave us in the dark, and we don’t know what the field looks like or what the rules of the game are.”
Ullrich, who worked at a power plant in his hometown of Lakeland, Fla., during the 1990s, recalls the industry’s “great uncertainty” during the period “about what was going to happen – how far deregulation was going to go. Nobody was building new power plants. If you look at the (EIA) data, you can see that excess capacity — the amount of generating capacity over expected demand — dropped through the decade.”
Ullrich developed an interest in the power business while interning at Lakeland Electric when he was a physics major at the University of Florida. Those summer jobs led to a full-time position at the utility after he graduated. He earned an M.B.A. as a part-time student before eventually leaving his job in 2000 to pursue a doctorate in finance at the University of Maryland. His industry knowledge and experience, he has discovered, definitely give him street cred with his students when he discusses industry issues in his finance classes.
His industry background has also made him more understanding and patient during an outage. “Almost all power outages are due to breakdowns in the transmission system, not to a lack of power.” He feels great empathy for the linemen who have to fix downed lines and face disgruntled customers. “The linemen have a very dangerous job and work in very bad weather conditions. Don’t be mad at the people who are trying to fix the problem.”
Ullrich, who sees his research specialty as a natural convergence of his power industry background and his academic training in finance, says the study of electricity markets and pricing is an emerging specialty field in finance. His other research projects have looked at risk management in the electricity industry and volatility in electricity markets.
Models must reflect price volatility
Because electricity is not storable in economically meaningful quantities, pricing methods used for derivative securities, such as options, forwards, and futures, cannot be used for derivatives written on electricity. Electricity prices behave differently than stock prices. “A large jump in electricity prices during the day is inevitably followed shortly by a reversal, or a return to the normal price level. Electricity prices also have intra-day patterns that vary by the time of year.” Finding good ways to price electricity derivatives is important because power plants are essentially options and “we need good methods for valuing power plants.”
Ullrich’s research has uncovered a significant problem with the data used in the mathematical models developed so far. Almost all the models use data sampled at daily, or lower, frequencies, he says. “But if you’re trying to value an option, such as a gas-fired power plant that has the flexibility to respond to intra-daily or higher-frequency prices, a model based on lower-frequency daily prices will certainly undervalue the option. To properly value these things, I need a model that’s built on the same time scale as the thing I’m trying to value.”
Ullrich is one of the first researchers to apply high-frequency electricity price data to study electricity markets. Using such data, he concluded that electricity markets in North America are, on average, just as volatile as those in Australia — in contrast to previous studies suggesting that Australian electricity markets are more volatile. The discrepancy was due to the other researchers using daily average prices, he says, “while I used the prices from which the averages were computed.”
The significance of “observation frequency,” he says, has not been emphasized in the literature. “For practical applications, the model should be fitted to data observed at the frequency that is relevant to the problem at hand.” Using daily prices would be appropriate, for example, for valuing a power purchase agreement for which settlement is based on average daily prices. “But such a model would not be well-suited to valuing an option, real or financial, that can be exercised hour-by-hour.”
Now is a great time to be studying electricity markets and the power business, Ullrich observes. Renewable energies and other green technologies are receiving greater attention, and climate and energy legislation, many believe, will transform America’s economic and industrial landscape. The power industry is deregulating, which requires a lot more business expertise. “When it was fully regulated, you didn’t need as many finance or business people as engineers to run it.”
His research, he hopes, will not only help the power industry manage risk more effectively and value plants more accurately but also contribute knowledge on how industries and investors in general respond to regulatory uncertainty as well as lead to policy prescriptions for regulators and lawmakers.