 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
Prepared Witness Testimony The House Committee on Energy and Commerce
Natural Gas Supply and Demand Issues
Full Committee on Energy and Commerce
Dr. Jeffrey R. Currie
Mr. Chairman and Members of the Committee, thank you for the opportunity to testify before you today about the short-term and long-term issues surrounding the natural gas market. My name is Jeffrey Currie. I am a Managing Director of Goldman Sachs, where I am the Senior Energy Economist. The views presented here today are my own and do not necessarily reflect the views of Goldman, Sachs & Co. The current shortage in the natural gas market is quite different from a normal cyclical shortage, and more dramatic action than simply allowing the market to function will be necessary to address the core problem, which is significant underinvestment in basic infrastructure. Public attention has been focused on the ability to grow natural gas supply. However, in this case, the underlying shortages in storage and transportation are the primary constraint on both supply and demand growth. The infrastructure in natural gas is so depleted that much of the adjustment has been and will continue to be in demand. Since demand is the quicker and lower-cost margin of adjustment, rather than supply, price spikes are likely to lead to demand destruction, which will quickly result in dramatic price declines. The much-needed investment in new infrastructure, however, has been and continues to be discouraged by poor returns that are exacerbated by an increasingly risky price environment. Since demand adjustments are not a long-term solution to the problem, shortages will develop again once demand recovers, creating a subsequent spike in prices. Further, these shortages in basic underlying infrastructure have prevented efficient use of existing supplies and efficient development of new supplies, which suggests that solving the basic supply problem will not, by itself, resolve the deliverability problems currently facing the natural gas market. The basic supply question of whether to open up areas to drilling or depend on LNG imports is a very important long-term issue. However, due to the current infrastructure constraints, even if there were significant surplus domestic natural gas (and there is in the Rockies), the market doesn't possess the pipeline capacity to transport it; and even if there were adequate pipeline capacity to transport this gas, which there is not, the market lacks the capacity to store it. Similar operational constraints also apply to potential LNG imports. As a case study, the winter and summer of 2001 demonstrate the economic impact of constraints on storage and pipeline capacity. That winter, severe shortages developed from a combination of cold weather and a lack of supply. Once inventories were exhausted, physical shortages turned critical, resulting in a massive price spike to $10.00/mmBtu that destroyed price-sensitive industrial demand to make room for essential heating demand. The loss in industrial demand was massive: a 20% permanent decline that resulted in the loss of at least 200 thousand manufacturing jobs (see Exhibit 1). Yet, the price spike also triggered a modest supply response, which when combined with the sharp drop in industrial demand, created a very large surplus of gas that took only six months to completely overwhelm the entire US natural gas infrastructure. By the end of the summer of 2001, surplus gas had nowhere to go, gas prices collapsed to under $2.00/mmBtu, and ultimately production had to be shut in (see Exhibit 2). The reason for this rapid reversal is straightforward economics - the industry did not possess the infrastructure to store or transport the surplus gas for a future supply shortage. When another shortage occurred only a year later in the winter of 2002/2003, the market had insufficient inventories to handle it. Looking forward from today, even if the industry filled storage to capacity by the end of this October, the inventory would still only cover 75% of all potential winter outcomes, leaving the market with a 25% chance of running into severe shortages before the end of next winter even under an improved supply outlook. Lack of storage capacity is the key determinant of natural gas price volatility These experiences of the last couple of years show that storage capacity is the key determinant of natural gas price volatility. Storage capacity provides the system with a buffer to supply and demand shocks by allowing it to run surpluses and deficits that smooth the normal cyclical swings in prices. As storage capacity has failed to keep pace with growth in demand over the past two decades, this buffer has shrunk relative to the size of the market, resulting in chronically higher-than-normal price volatility. In the 1980s, we had about 1,400 bcf of storage beyond that which is necessary to operate the system and deal with winter demand swings. This storage represented about 26 days of forward consumption, a significant shock absorber that generated relatively stable natural gas prices. Today, we have only 330 bcf of storage beyond what is necessary to run the system, which at today's higher demand levels is only 6 days of forward consumption. In response, price volatility has exploded to nearly three times the historical average (see Exhibit 3). Thus, fairly small deficits or surpluses will cause the market to move from full to empty and from $2 to $10/mmBtu or back in a relatively short amount of time. Poor rates of returns have resulted in underinvestment in infrastructure The broader question is, "Why has storage capacity and related infrastructure failed to keep pace with demand?" The answer in its simplest form is that a combination of regulation, taxes, and direct market intervention have made the return on capital in the energy industry a breakeven proposition at best and have made investing in the downstream (transportation, storage and other aspects of the infrastructure) distinctly unprofitable. The market has responded by not providing the capital to expand, and the net result is the capacity constraints that you see today. If you look at the industry as a whole during 2001, a year which posted the highest annual gas prices on record, and saw historically high energy equity valuations during the 1H2001, the industry was not even valued at the cash that had been invested into it, hardly a compelling return. Worse, if we exclude the super majors, the rest of the gas supply, transmission, and distribution industry was actually valued at only 73% of the cash invested (see Exhibit 4). It is hardly surprising that the market has not supplied sufficient additional capital to meet current demands. If we look deeper into the numbers, the lack of investment in basic core infrastructure (storage and transportation) becomes even clearer. E&P, the drilling part of the business, has earned a 5.6% return on assets on average over the last three years, while distribution and transmission, the infrastructure part of the industry, has earned only a 2.4% return on assets (see Exhibit 5). This return on assets for downstream companies is considerably below the 5.0% return on assets earned by the broader S&P 500 index in the second half of the 1990s. The reality of modern capital markets is that only industries with significant positive returns on cash invested above the cost of capital attract new capital. If you compare return on cash invested across industries over the last decade for companies in the S&P 500, the reason for today's energy shortages become quite transparent. Utilities and energy companies managed to produce slightly less than a 9% return on cash invested while the rest of the market produced returns on cash invested of 12.5% and above (see Exhibit 6). It is hardly surprising that most of the investment activity has occurred elsewhere, stressing our energy infrastructure to its limits. Controlled "deregulation" increases risks on poor returns Worse, the risks associated with these poor returns have increased significantly since the mid-1990s due to "deregulation" and "environmental rules." Clearly, the introduction of competition over the last decade has increased the risks associated with investments in energy infrastructure. In natural gas storage and transmission, controlled deregulation as opposed to true competition has dramatically increased risks (primarily volume risks). However, the rates of return on these assets have not risen over the last decade to compensate for the higher risks. Rather, the rates of return have fallen, which makes the situation worse on a risk-adjusted basis. Further, following "deregulation," the rates of return were supported primarily through cost reduction, as the emphasis in the industry shifted from reliability to efficiency, i.e. through getting rid of the excess. This is all too apparent in the drop in transmission and gathering pipeline capacity that was deemed "excess" during the 1990s (see Exhibit 7). To internalize these risks, the industry in the past has relied upon long-term forward contracts or some form of vertical integration. Current regulations, however, discourage both of these forms of risk management, as the emphasis is placed on the use of spot prices and the transparency they provide to both consumers and producers. This spot price transparency is very effective in providing market signals for efficient drilling and consumption patterns, which are relatively low-capital intensive activities. However, for more capital intensive and longer lead-time activities, such as building infrastructure, a spot market price signal is a lagging indicator of an investment that should have already been made. Instead, forward contracts of sufficiently long duration are needed to internalize the risks and induce the needed investment in advance of shortages. Further, current regulations require any long-term contracts to build infrastructure to have such a high subscription rate, near 80%, that excess capacity will rarely be built, which reinforces the underinvestment problem. Policy needs to create reserve capacity that market forces are failing to generate The paradox of the current situation is that the underinvestment in infrastructure by the market is the correct economic outcome given the poor rates of return, as the best use of capital is in other industries where the rates of return are higher. The market solution is not concerned with volatility, but rather the expected rate of return. This solution only leads to new infrastructure when it is absolutely needed, which is usually too late. Just look at the only large infrastructure projects of the last several years - the Alliance, Kern River, and Gulf Stream pipelines - projects brought about by extreme pricing. However, reserve or excess capacity should be viewed as a public good, just like a road, where markets fail to find a solution. This inability of the market to provide adequate incentives for investment in reserve infrastructure is where the market fails and why more dramatic action is required. Further, the current market and regulatory structure reinforces this price volatility as it emphasizes efficiency over reliability. Accordingly, the aim of policy should be to reduce the price volatility through creating excess capacity without significantly sacrificing the efficiency and transparency of a market-based system. Forcing excess capacity through regulation has not been met with much success in the past. Before the 1980s, regulatory practices emphasized reliability by requiring pipeline companies to demonstrate sufficient capacity to serve additional customers before projects would be approved. To internalize the risks of such ambitious projects, 30-year long-term contracts with regulated price caps were often used. These price caps were fixed and ultimately led to significant market distortions, as the market could not clear properly. The stranded costs generated during this regulatory period have been estimated at $80 billion in 2002 dollars. Interestingly, the costs to consumers due to increased volatility in the post-"deregulation" period are not much smaller. Since 1995, these costs, measured as the cumulative difference between the price paid and marginal cost of production is near $75 billion in 2002 dollars, nearly the same as stranded costs generated from the regulatory period, and this does not include the costs of the California crisis and the long-term loss of manufacturing activity. Further, with the cost of an arctic pipeline estimated at $10 billion, these costs would have paid for new infrastructure and then some. What this suggests is that transportation and storage assets may be thought of as public goods and could be treated just like a freeway or toll road. The US energy consumer would have most likely been made better off had the government taxed natural gas prices and used the proceeds to build infrastructure, just as it taxes gasoline to build roads. The key issue is to create excess capacity that market forces are failing to generate. This would dramatically reduce price volatility, investment risk, and create a more conducive environment for demand growth. The lack of infrastructure is a limiting factor on economic growth. Energy is rapidly becoming a major limiting factor on economic growth. If the core energy infrastructure in the United States does not improve, energy crises are likely to become progressively more frequent, more severe, and more disruptive of economic activity. Without significant new investment, each crisis further damages the system by permanently destroying the price-sensitive demand that serves as a pressure valve and by giving companies incentives to stress existing facilities to meet excess demand, leading to accidents and capacity losses. The long-term consequences of either allowing infrastructure to remain inadequate or sacrificing environmental concerns in the name of economic expediency are unacceptable. Finding a "workable" solution will require imagination and flexibility from both a market and policy perspective. Economic solutions depend on diversification of risk and flexibility of response, both of which are lacking under the current market and regulatory structure.
Exhibit 1: Demand destruction has been concentrated in the manufacturing sector. 1,000 of manufacturing jobs (left axis); $/mmBtu (right axis)
Source: Bureau of Labor Statistics and US Department of Energy (DOE).
Exhibit 2: Lack of storage capacity constrains the market's ability to adjust. Bcf
Source: DOE and Goldman Sachs Research.
Exhibit 3: As reserve capacity has fallen relative to the size of the market, price volatility has increased. Days of forward coverage (left axis); $/mmBtu (right axis)
Source: American Gas Association (AGA), DOE, and Goldman Sachs Research. Exhibit 4: The natural gas industry is valued at less then the cash that was originally invested
Source: Goldman Sachs Research calculations using Compustat data. Exhibit 5: Poor downstream returns have generated the current infrastructure constraints. Percent return on assets
Source: Goldman Sachs Research calculations using Compustat data.
Exhibit 6: Energy sector returns have underperformed relative to most other sectors
Source:Goldman Sachs Research calculations using Compustat data. Exhibit 7: Excess capacity has declined following deregulation as the emphasis shifted to efficiency
Source: Goldman Sachs Research calculations using Compustat data.
|
|||||||||||
|
