Smith suggests that one policy implication is to offer leases that cover larger areas-large enough to completely encompass a geological formation.

The following is a summary of the paper ""Rational Plunging and the Option Value of Sequential Investment: The Case of Petroleum Exploration" which is forthcoming in the Quarterly Review of Economics and Finance.

Though Republicans and Democrats find ways to make more drilling possible by opening up previously restricted areas, the actual decision by a firm whether to drill or not matters too. The policy “Drill, baby, drill” may have a few unexpected nuances that petroleum and gas exploration firms control. New research* by Professors James Smith and Rex Thompson of SMU Cox explores the question of why, rather than diversifying their exploration efforts, oil and gas firms tend to “plunge” and concentrate their portfolios of assets in like exploration prospects such as a “play” that contains common geological or risk factors.

When bidding on leases of tracts to replenish their exploration portfolios, firms choose to concentrate their assets more than diversify. This has implications for policymakers grappling with the right formula to incentivize new drilling. Thompson says, “Why would firms tend to concentrate their portfolios of assets? The learning that comes from sequential prospects is an offset to the gains of diversification. There’s tension between diversifying and the learning that comes when you drill the first hole. You may learn something beneficial; if not, you might walk away and save on drilling costs. This is particularly true where the value of learning can’t be captured by the driller.”

A key to the drilling debate could easily be missed—why firms choose their new prospects and what government could do to smooth the process. Smith suggests that one policy implication is to offer leases that cover larger areas—large enough to completely encompass a geological formation. He says, “Few other countries in the world offer such tiny parcels. Why do we? Perhaps it is to let the "little guy" have a chance, but that argument is pretty outdated when it comes to the oil business.” Referring to the research findings, Smith relates that when the decision regarding tract size was made in the 1950s, the extent of information spillovers (the learning aspect) was not fully appreciated.

Background
The authors wanted to determine whether a portfolio of exploration prospects, an energy firm’s real assets, would reveal diversification like a portfolio of financial assets such as stocks, bonds, real estate, and so on. Thompson explains their thinking: “With stock portfolios, if you invest in a number of bank stocks, and they decline, you can’t unwind your positions fast enough to beat the losses. You’re in it, and it happens simultaneously. So learning in a traditional finance portfolio isn’t one of its features. You get outcomes, but can’t act on them. With oil and gas prospects what you learn from early drilling experience influences your later behavior, which is the important difference,” Thompson said.

“We began to see that the traditional financial model of portfolio diversification didn’t explain what’s happening in the real world with portfolios of real assets. Many R&D decisions mirror this plunging strategy which includes the learning aspect. You see this in biotechnology and pharmaceuticals as they choose to specialize in certain drugs,” Thompson mentions.

In the research, the authors analyzed the process of how firms assembled their portfolios of exploration prospects. A few facts help begin the framing of the process. On average, the net value of U.S. discoveries (gross minus development costs) is roughly 15 times the cost of exploration. Said another way, an estimated rate of return calculated by the late Stephen McDonald of the University of Texas at Austin (1970) is 14.5%, whereas Walter Mead of the University of California at Santa Barbara (1983) estimated wildcat exploration in the Outer Continental Shelf at 12.3%. Thus exploration is a profitable business; all the more profitable if the percentage of dry holes is managed carefully. This is where learning comes into play.

Since 1954, the U.S. Government has periodically auctioned rights to explore for petroleum on designated offshore tracts located on the Outer Continental Shelf (OCS). A typical auction (lease sale) includes numerous tracts; each company must select individual properties on which to bid. Exploration rights on more than 24,000 tracts have been awarded via this process. The total value of all bids tendered since 1954 exceeds $135 billion, of which the high (winning) bids amount to some $64 billion (unadjusted for inflation). Initially, the OCS auctions were conducted by the U.S. Geological Survey (USGS), but administrative responsibility passed to the newly created Minerals Management Service (MMS) in 1982. These auctions have long been, and remain today, an active and economically significant market which is used regularly by oil and gas companies to assemble and replenish their exploration portfolios.

The rich data set of government lease sales used by the researchers ranges from some of the very smallest, privately-held companies to the large multinational firms that dominate the petroleum industry. Their study focuses on five specific lease sales that are among the largest lease sales ever to have been held. The five sales all took place between June 1973 and October 1974, at the very height of concern over future petroleum supplies; and all five were categorized as “wildcat” sales—which means that no companies had yet been given a chance to conduct test drilling in the vicinity of these tracts. Therefore, no participants had accumulated much proprietary information of a type that would be difficult to identify or control in the analysis.

The playing field for exploration in many regions of the world will never be level again. Smith mentions, “Today, after forty years of drilling in government-owned lands in the Gulf of Mexico, many firms hold proprietary knowledge gleaned from their individual exploration efforts. In the 1970s, no one held any proprietary information of that type, at least not regarding the tracts in our sample.” He explains that companies acquire information about offered areas through consortia, pooling resources to fund seismic surveys and then sharing the results among themselves and the government. The pending Congressional energy bill that would lift a 27-year-old moratorium that has prevented drilling along the Atlantic and Pacific coasts and certain areas of the Gulf of Mexico would restore exploration opportunities similar to what existed in the wildcat sales of the 1970s.

Overall, a total of 582 tracts drew bids in the five lease sales studied by Smith and Thompson. The number of participants (bidders) varied between 51 and 82 per sale, and the average participant tendered 17.3 bids per sale. The 582 tracts were spread across 193 distinct geological structures, giving on average 3 tracts per structure. A set of tracts associated with a common geological structure is referred to as a “group” of related tracts. The number of such groups varies between 11 and 65 per sale, and the number of tracts per group varies between 1 and 33. With this array of tracts on offer, participants in each auction could have pursued either a concentrated or diversified portfolio of exploration prospects, according to their preference.

Learning Matters
The large majority of firms assembled portfolios that tended to be geologically concentrated on just a few groups rather than diversified. Companies attempt to exploit the value of “information spillovers,” or the learning involved, by pursuing prospects that are geologically related rather than diversified. The authors find that the portfolios assembled by private firms are even more concentrated than the portfolios of public firms.

Under conditions typical of U.S. exploration, the incentive to plunge into concentrated holdings is greater for risk-averse companies than for risk-neutral companies. Thompson notes, “Publicly-traded companies such as Exxon and BP tend to act in a more risk-neutral way, partly because their shareholders are diversified beyond Exxon. And while Exxon and other large publicly-traded firms tend to plunge, this tendency is even stronger among privately-held companies, which are presumed to be more risk averse.

Many oil firms are “accused” of owning drilling leases but not drilling on them, appearing on the surface to frustrate energy challenges. Smith says that companies delay drilling their leases for various reasons. “For example, a company might acquire a lease that would require either an improvement in price, or a breakthrough in offshore technology, to be developed profitably,” he notes. “Sometimes drilling on other tracts is delayed pending the outcome of adjacent drilling--for all the reasons we explore in our paper. Some companies delay drilling, especially in times like the present, when daily rig rates and other costs have escalated steeply due to the surge of activity triggered by $100 oil.”

Implications
Any investor in assets that may be exploited sequentially faces an inherent tradeoff. Does he/she create a portfolio with assets whose returns are correlated, making total return swing more widely? Or does he/she extract the value (the learning) from the options that naturally arise due to the interdependence among assets? Learning allows an investor to change behavior, which can mitigate against future risks. The authors suggest that the value of the options increases according to the strength of the dependence among assets. Thompson provides an analogy: “Imagine a spectrum of risk-averse to aggressive entrepreneurs. The more risk-averse will see two benefits from learning potential, both increased average return and decreased risk. The aggressive see only the potential to increase return. Thus the learning option means less to the risk-takers, and they won’t pay as much for it. The risk-averse firm will pay a premium for a tract that concentrates its portfolio if the tract also generates a learning option.”

Thompson further qualifies his angle, “If our research is extended to include other types of R&D, we wouldn't expect to see firms biting off huge chunks of related risks in all situations, or plunging—they will plunge only where learning provides enough benefit to offset the lack of diversification.”

In terms of the heated energy debate loaded with misinformation, these new insights may afford a head start in the right policy direction.

* “Rational Plunging and the Option Value of Sequential Investment: The Case of Petroleum Exploration” by oil and gas economics professor James L. Smith and finance professor Rex Thompson of SMU Cox is forthcoming in the Quarterly Review of Economics and Finance.