Printer Friendly

A transaction cost model of contract choice: The case of petroleum exploration.

Abstract

The growing literature on transaction costs posits that the structures of contracts involving exchange under uncertainty are influenced by the costs incurred by the contracting parties prior to, as well as after, a contract is signed. This research investigates the contractual responses to the substantial uncertainty attending the exchange of rights to underground petroleum deposits. It develops a transaction cost model to explain the payment structure found in these contracts. The model identifies the major transaction costs associated with the payment types used in oil and gas exploration contracts, including ex ante measurement costs and ex post production inefficiencies, and explains their effect on contract structure. Testable implications concerning variations in the payment structure of petroleum exploration contracts are generated and tested using data from private oil and gas mineral rights leasing contracts in four western states. The study has direct public policy significance in that it delineates the implications of different payment structures of oil and gas leasing contracts. These implications can be used to evaluate proposals to reform federal oil and gas leasing policies. In addition, while there has been considerable analysis of federal offshore oil and gas leasing contracts, there has been a dearth of research on private onshore oil leasing practices. This study helps to fill this empirical void. (JEL Q20)

Introduction

Contracts in which part of the payment structure is conditional on the ex post value of a good or service being exchanged are commonly found when the exchange of property rights occurs under substantial uncertainty about the value of the good or service being exchanged. A standard explanation for contracts containing conditional payments focuses on the risk sharing and incentive properties of different contractual arrangements. For example, in well-known principal-agent models by Grossman and Hart [1983], McAfee and McMillan [1986] and others, conditional and fixed payments impose risk differentially and provide differing incentives for input use by the contracting parties. An alternative explanation for these types of payment structures focuses on the effects of transaction costs. Models by Cheung [1969], Williamson [1979], Barzel [1982, 1987], Leffler and Rucker [1991], Allen and Lueck [1999], Black [2000], and others demonstrate that the structures of contracts involving resources with uncertain value are influenced by the transaction costs incurred by the contracting parties prior to, as well as after, a contract is signed. The purpose of this study is to extend the growing transaction cost literature by developing a transaction cost model of oil and gas mineral rights leasing contracts.

When the rights to potential petroleum deposits are exchanged, compensation consists of both lump sum and conditional payments. These payment types differ with respect to the type, size, and timing of transaction costs incurred by the contracting parties. Under the considerable uncertainty attending petroleum exploration, the existence of pre-sale and post-sale measurement, enforcement, negotiation, monitoring, and other transaction costs can constrain the gains from different contract structures. The choice of payment structure is seen as an optimizing decision under the constraints of these transaction costs.

Property Rights in Petroleum Development

In the United States, petroleum is treated in property rights law as a mineral and ownership of land generally carries title to the subsurface mineral resources directly beneath. This is in contrast to the assignment of property rights in most nations, where title to subsurface minerals resides with the state as sovereign. In the U.S., ownership of an in-situ petroleum deposit is secured through the possession of mineral rights. Mineral rights can be exchanged like any other private good. The landowner may sell or lease the rights to subsurface minerals separately from the rights to the land surface. The instrument through which a transfer of rights to subsurface petroleum resources takes place is the oil and gas mineral rights lease. Such leases typically delineate the specific rights transferred, the duration of the transfer, and the form of payments. The rights specified include the rights to any oil and gas produced from wells drilled into potential petroleum deposits beneath the acreage covered by the le ase. The duration of the lease, called the primary term, is generally in the range of three to five years. (1) In these lease contracts, payments take the form of some combination of fixed and conditional payments. The conditional payment component consists of royalty payments based on a percentage of the gross revenue from the future sales of any oil and gas production from the lease. (2) In contrast, the fixed payment component, termed the bonus payment, is an up-front, lump-sum payment similar to those found in most market transactions. Bonuses are fixed per-acre payments based on the amount of surface land covered by the lease. As such, they are not contingent on any production from the lease. This fixed payment component of the overall payment structure is contrasted with the conditional nature of royalty payments.

The payment schedules in petroleum leases vary in their emphasis on bonus versus royalty payments. A contract that consists of pure bonus payments and no royalties is equivalent to an outright sale of property rights for the duration of the lease. On the other hand, under a pure royalty contract, payments are conditional on the unknown value of the petroleum produced in the future. In the model developed here, variations in the amounts of bonus versus royalty payments result in differences in the transactions costs incurred by the contracting parties. The types and degree of transaction costs also vary depending on the degree of uncertainty regarding the presence and value of subsurface petroleum deposits. In the next section, the effects of transaction costs associated with contracting under uncertainty are delineated.

Transaction Cost Model of Contract Choice

The transaction cost model developed here is based on the fact that the types of transaction costs associated with fixed payments are often different from those associated with conditional payment provisions. Transaction costs associated with fixed payments stem primarily from efforts on the part of buyers to engage in pre-sale measurement. In the context of oil and gas leasing, it is shown here that there is a positive relationship between the degree to which bonus payments comprise the payment structure and the incentive on the part of oil producers to engage in pre-sale measurement. It is also shown that with competitive bidders, the mineral rights owner ultimately bears these measurement costs. As a result, the mineral rights owner has an incentive to try to reduce the amount of pre-sale measurement on the part of oil and gas firms. One method of reducing pre-sale measurement costs is through the use of conditional payments. While reducing measurement costs, however, conditional payments generate other ty pes of transaction costs. The use of royalties in oil and gas leasing leads to higher losses from inefficient production, speculation, and contract negotiation than do contracts in which the payment structure consists of only bonus payments.

Transaction Costs and Fixed Payment Contracts

The uncertainty regarding the value of any tract of land in terms of petroleum production provides incentives for potential buyers to expend resources in efforts to increase the amount of information regarding resource value prior to making offers for mineral rights. By engaging in some degree of pre-sale measurement, potential buyers can improve the terms of their offer so as to increase their expected profit.

The incentives to engage in pre-sale measurement on the part of buyers when there is ex ante uncertainty can be illustrated using a standard auction model. Consider the case where oil and gas producers compete for leases by making bonus payment bids to mineral rights owners and where leases go to the highest bidder. Where several oil and gas producers are competing for the mineral rights to a prospect, and where the mineral rights are held by a relatively small number of mineral rights owners, the situation is analogous to that of a first-price, sealed-bid, common-value auction. In these auctions, potential buyers submit sealed bids, the highest bid wins the auction, (3) and the object being sold has a "single objective value, such as the amount an antique is worth on the market or the amount of oil actually lying beneath the ground" [McAfee and McMillan, 1986, p. 705].

Competitive leasing on federal lands in the U.S. closely approximates the features of first-price, sealed-bid, common-value auctions. As a result, this type of model is used in a number of studies of federal leasing of outer continental shelf (OCS) tracts. These studies include those by Hendricks and Porter [1988], Ramsey [1980], and Reece [1978, 1979]. The leasing process on private lands in the U.S. is not statutorily competitive, as is OCS leasing. However, numerous major and independent oil producers simultaneously compete for available leases in all but the most frontier areas. Therefore, the adoption of a first-price, sealed-bid, common-values auction model with competitive bidders is taken to be a reasonable approximation of onshore leasing of private lands.

Consider the case where a number of oil and gas producers bid for the rights to a potentially valuable mineral lease. All bidders are identical in the sense that they share the same initial estimate of the size of potential petroleum deposits associated with the lease. This estimate is based on publicly available information such as geophysical well logs, production information, and published geological information. Models of oil and gas lease auctions by Gaskins and Teisberg [1979], Englebracht-Wiggans, Milgrom, and Weber [1983], Milgrom and Weber [1982], and others show that a bidder who possesses only public information has zero expected profit, but that supplementing publicly available information with private information raises a bidder's expected profits. Indeed, studies of oil and gas leasing on OCS tracts by Mead, Moseidjord, and Sorenson [1985], Hendricks and Porter [1988], Hendricks, Porter, and Tan [1993], and Hendricks, Porter, and Wilson [1994] show that bidders possessing private information hav e higher returns to bidding than those who do not. In addition, as noted by Gaskins and Teisberg [1979], the incentive for prospective bidders for oil and gas leases to acquire private information is further supported by the observation that a great deal of exploration activity takes place prior to OCS lease auctions.

The potential for informational rents in competitive auctions poses questions regarding if, and how, these rents are dissipated. With an endogenous number of bidders, a competitive equilibrium requires that these returns must either be dissipated by bidders or captured by the seller. (4) Although Hendricks, Porter, and Tan [1993] and Hendricks, Porter, and Wilson [1994] develop models in which informational rents can be partially captured by sellers through the use of reservation prices, such schemes are difficult to execute and rarely used in oil and gas leasing. The most likely explanation for the dissipation of informational rents is that of dissipation by bidders. In the case of oil and gas exploration, this arises due to costly pre-sale measurement by oil and gas firms prior to bidding. The process of discovering a new field often takes from several months to several years and involves large expenditures consisting primarily of the costs of gathering and evaluating geological information, performing seis mic surveys, acquiring leases, and drilling exploratory wells.

These pre-sale measurement costs are crucial in explaining the structure of oil and gas leasing contracts. However, the effects of such costs have been ignored in most models of first-price, sealed-bid auctions. Notable exceptions include studies by French and McCormick [1984], Johnson [1979], Leffler and Rucker [1991], and Ramsey [1980]. The model of bid preparation costs developed by French and McCormick [1984] (5) demonstrates that when bidders compete for a mineral lease offered by a non-competitive owner, the pre-sale measurement costs incurred by all bidders are borne by the mineral rights owner. The intuition for this result is analogous to the case where sellers with a perfectly inelastic supply curve for a product bear all the burden of a tax. In the case of mineral rights leasing, the owner pays for the pre-sale measurement on the part of bidders through a reduction in the winning bid. As a result, sellers have an incentive to reduce the measurement costs for their particular tract. As noted by Leff ler and Rucker [1991], a decrease in measurement costs induces a less than proportional increase in the number of bidders, causing a decrease in total measurement costs and an increase in the seller's revenue.

One way for sellers to reduce the amount of pre-sale measurement on the part of buyers is to provide information about resource value to potential bidders. This is indeed done by the federal government to increase revenues. For example, the United States Geological Survey provides reserve estimates for offshore tracts prior to lease sales. On the state level, requirements that well logs be made publicly available can also be viewed in this light. This is almost never done on privately held land, however. A more viable approach for private landowners to reduce pre-sale measurement costs is to use share contracts in which payment is conditional on the realized value of the good. As noted by Barzel [1982], because successful bidders pay according to the actual value of the resource produced, the use of share contracts reduces the incentive on the part of buyers to engage in pre-sale measurement. As a result, measurement costs incurred by sellers are reduced. This type of payment structure is examined in the foll owing section.

Transaction Costs and Share Payment Contracts

If ex ante measurement costs were the only source of transaction costs, it would be expected that all resource sales would take place with payments being strictly on a share basis [Leffler and Rucker, 1991]. In the case of oil and gas leasing, this would consist of mineral rights being transferred to the bidder offering the highest royalty rate. Share and other conditional payment provisions, however, entail their own set of transaction costs. These costs differ from those generated by fixed-payment contracts and include losses due to the incentive to underproduce the resource, losses due to speculation, and higher negotiation and enforcement costs.

Costs Due to Underproduction

First, and probably most importantly in the case of petroleum production, share contracts provide an incentive for the oil producer to extract less oil than under fixed payment contract. With increasing marginal costs of extraction, the buyer of the resource rights will under-produce the resource from the perspective of the seller. The profit-maximizing amount of extraction on the part of the buyer will be less than the level of extraction that maximizes the joint profits of the buyer and seller of the mineral rights. The seller can mitigate this only through costly enforcement actions.

In the context of petroleum production, the presence of royalties will cause an oil and gas producer to produce less than when the payment for the mineral lease consists only of bonus payments. This results from the incentive on the part of the producer to prematurely abandon the reservoir in the presence of royalties. The effect of royalty payments on oil and gas field abandonment is qualitatively discussed by several authors, including Leland [1978], Leland, Norgaard, and Pearson [1974], Mead et al. [1985], and Reece [1979]. Black [2000] demonstrated this effect theoretically by incorporating geophysical parameters of petroleum production into producers' wealth maximizing decision problem. He also developed an expression for estimating the magnitude of the underproduction and showed that underproduction is a function of the royalty rate used in the payment contract and geological parameters such as the field production decline rate and the economic terminal production rate. These losses from underproduction can be substantial. Black [2000] estimated that with a standard 12.5 percent royalty rate, underproduction losses amounted to 1.5 percent to 6.2 percent of the total cumulative production for the fields in his sample. With higher royalty rates and lower production decline rates, these losses can be much higher.

Although such costs due to inefficient production are likely to be the lion's share of transaction costs due to the use of share payments, two additional types of transaction costs generated by conditional payment structures are discussed in the literature. Mead et al. [1985] and Rucker and Leffler [1988] discuss potential losses to the seller due to speculation and non-development when royalties comprise a major part of contract payment provisions. In addition, Cheung [1969] found that share contracts result in higher negotiation costs than fixed payment contracts. These are discussed briefly below.

Costs Due to Speculation

In addition to forgone revenues due to premature abandonment of producing fields, royalty bidding can result in forgone revenues by discouraging exploration and initial lease development. As discussed earlier, the primary term for most oil and gas leasing contracts ranges from three to five years. The term of the lease is extended beyond the primary term if production occurs from the lease. Otherwise the lease expires at the end of the primary term. Thus, oil and gas leases have characteristics of option contracts. With bidding taking place on the basis of pure royalty payments, the winning bidder may secure the right to drill and produce oil for the duration of the primary term without making any payments.

As Mead et al. [1985] point out, oil producers can submit high royalty bids in order to win leases and may not intend to develop them unless oil and gas prices unexpectedly increase. They note that with experimental offshore leasing programs conducted by the Department of Interior in which bidding occurred on a pure royalty rate basis, the resulting high royalty rates led to a marked decrease in exploration on leased tracts with a commensurate decrease in the number of commercial finds. Due to these results, the program was discontinued. Similar problems on North Sea tracts, cited by Dam [1976], led the British government to selectively remit royalties in order to encourage the development of marginal fields. Rucker and Leffler [1988] developed supporting evidence for similar speculative incentives to default on timber harvesting. Thus, in addition to decreasing the amount of production from discovered fields, high royalty rates reduce the probability of commercial production in yet-to-be explored areas.

Costs Due to Negotiation and Monitoring

The marginal costs of negotiating contracts and monitoring and enforcing contract provisions are likely to be higher under royalty contracts than under bonus payment contracts. The potential for speculation losses often lead to the inclusion of contract terms designed to increase the likelihood that the oil producer will undertake exploration and production efforts. In addition, it is in the landowner's interest to negotiate incentive clauses to increase production on the part of the producer due to the underproduction effect described above. The monitoring and performance stipulations included in royalty contracts can be complex to negotiate and enforce. In contrast, these transaction costs are relatively low for contracts consisting solely of fixed payment provisions.

The foregoing discussion coincides with Cheung's [1969] claim that negotiation costs in agricultural contracting are higher for share contracts than for fixed payment contracts. With payments occurring on the basis of fixed payments, the seller receives payment regardless of any production occurring from the leased acreage. Because no further payment arises from production, the incentive on the part of the mineral rights owner to engage in costly efforts to ensure development of the lease is much reduced. With share contracts, both seller and buyer have an incentive to contribute to the decision-making with regard to production, resulting in a larger degree of coordination and negotiation.

Empirical Test of the Transaction Cost Model

Transaction cost analysis explains the choice between royalty and bonus payments as being determined by the effects of the transaction costs generated by each payment type. On the one hand, factors that tend to increase the costs of pre-sale measurement on the part of bidders will provide an incentive for the seller to favor royalty payments over bonus payments. On the other hand, factors that tend to increase the costs of underproduction, speculation, or negotiation will provide an incentive for the seller to favor bonus payments rather than royalties. Contracts for oil and gas mineral rights contain both fixed and conditional payment provisions. There is a great deal of variability in the degree to which such contracts emphasize bonus versus royalties. The transaction cost model developed here forms the basis for testable predictions about the payment structure of these contracts.

Testable Implications

The first of these predictions concerns the impact of resource value uncertainty on contract structure. As discussed earlier, several studies show that there are positive returns to acquiring information prior to bidding in cases where resource value is uncertain, and that these returns increase with the degree of uncertainty. (6) With increased resource value uncertainty, there is more incentive for potential bidders to engage in costly pre-sale measurement. In the case of oil and gas exploration, such measurement consists of geological and geophysical surveys, the costs for which can easily run into millions of dollars for even relatively small areas. In order to reduce these costs, sellers have an incentive to increase a contract's reliance on conditional payments rather than on fixed payments. In other words, an increased degree of uncertainty regarding resource value will result in the optimal payment structure having a higher royalty payment component relative to the fixed payment component. Thus, the t ransaction cost model predicts a positive relationship between resource value uncertainty and the use of royalty payments.

To test this prediction, differences in the contract payment structure are related to uncertainty in resource quality. Two proxies of uncertainty are used. The first is the distance from lease locations to the nearest producing fields. It can be shown that with an increase in the distance from the lease location to the nearest producing field, there is an increase in the variance of the a priori distribution of resource value, and a commensurate increase in the uncertainty of resource value. The variable used to capture this source of uncertainty is termed DIST. Transaction cost analysis predicts a pattern of decreased emphasis on royalties relative to bonus payments as one moves from frontier areas toward known fields. Because there is a decrease in the uncertainty of resource value, there is a commensurate decrease in the incentive on the part of bidders to engage in pre-sale measurement. There is, as a result, less incentive on the part of sellers to use royalty payments as a means to reduce the measuremen t costs they incur. Thus, transaction cost analysis predicts that the royalty component of the payment structure will decrease relative to the bonus payment component with a decrease in DIST.

The second measure of uncertainty is the difference in the size of producing fields proximal to a given lease location. An increase in the variation in the size of nearby producing fields increases the variance of the prior distribution of resource value and results in a commensurate increase in uncertainty. This proxy for resource uncertainty, VAR, is measured by the variance of the size of the two producing fields closest to each lease. The measure of field size used to construct VAR is average annual production and is calculated by dividing cumulative oil production by the number of years of production for each field. This more accurately depicts field size than the use of cumulative production or annual production alone. (7) The transaction cost model predicts a pattern of increased emphasis on royalties relative to bonus payments as the value of this variable increases.

The predictions of transaction cost analysis in relation to the VAR variable are similar to those related to the DIST variable. A low value of VAR implies less resource value uncertainty and the associated contracts are predicted to have low royalty payment components relative to the bonus payment components. With lower uncertainty of resource value, there are lower returns to pre-sale measurement on the part of buyers and less motivation on the part of sellers to use royalties to reduce pre-sale measurement. Thus, transaction cost analysis predicts a positive relationship between VAR and the royalty payment component of the total payment structure.

An additional factor likely to affect the incentive on the part of buyers to engage in presale measurement is the size of nearby fields. With an increase in the expected discovery size of oil and gas fields, there is an increase in the incentive to engage in pre-sale measurement. This will increase the incentive on the part of sellers to use conditional payments in order to decrease the amount of such measurement by buyers. A variable, FIELDSIZE, is constructed by noting the size of the producing field nearest each lease. As with the variable VAR, size is measured in terms of average annual production. Thus, the transaction cost model predicts a positive relationship between FIELDSIZE and the emphasis on royalties in contract payment structure.

Further predictions are forthcoming from the transaction cost model. These predictions concern the relationship between the payment structure of oil and gas leasing contracts and factors affecting the costs of pre-sale measurement and negotiation. One factor affecting the costs of pre-sale measurement is the size of the firm leasing the mineral rights from the landowner. There is some evidence that small firms have lower costs of performing seismic and other types of exploration than major oil companies. Under this assumption, then, an increase in firm size leads to more costly pre-sale measurement and increases the incentive for the seller to use conditional contracts in order to reduce the amount of pre-sale measurement. Using FIRMSIZE as a measure of the size of the firm obtaining the mineral rights from the landowner, transaction cost analysis predicts a positive relationship between FIRMSIZE and the size of the royalty component relative to the bonus payment component of the payment structure.

One variable is generated for the purpose of testing the effect of negotiation costs on contract structure. Where more parties are likely to be involved in negotiations, negotiation costs will be higher, ceteris paribus. Transaction cost analysis predicts that, under this assumption, bonus payments are more likely to be used where the number of parties involved in negotiation is high in order to reduce the costs of contract negotiation and enforcement. A proxy for the number of parties involved in contract negotiations is constructed and termed PARTIES. A negative relationship between the PARTIES variable and the royalty component of the payment structure is predicted by the transaction cost model.

Finally, one variable is generated to account for the fact that high royalty rates may result in actions on the part of oil producers to delay exploration and production activities. Recall that high royalty rates can discourage exploration and development activities. This leads to forgone revenues to sellers. To counter this effect, sellers may specify shorter lease terms for leases with high royalty rates. Therefore, transaction costs analysis predicts a negative relationship between the length of the lease term, TERM, and the royalty component of the payment provisions.

The following section describes the process by which the predictions of the transaction cost model are tested using data from a sample of oil and gas exploration leases from four Rocky Mountain petroleum-producing states.

Empirical Analysis Data Description

Data on payment provisions and other aspects of oil and gas leasing contracts were solicited from oil producing firms in Montana, North Dakota, Wyoming, and Colorado. Two principal means were used. Detailed survey instruments were sent to major and independent oil and gas producing companies. In addition, some proprietary leasing records were obtained from oil producing firms. These data provided information about the payment structure of leases, including the amounts of bonus and royalty payments, lease dates and duration of leasing contracts, lease location and number of acres leased, characteristics of the contracting parties, and other contract provisions. In all, information on 140 recent mineral leasing contracts was obtained and used in the empirical analysis.

The data set indicates substantial variation in the payment structure. Bonus payments range from $1.00 per acre to $202 per acre, with a mean of $86.18 per acre and a standard deviation of $64.16 per acre. The royalty rates reported a range from 12.5 percent to 25 percent, with a mean of 16.07 percent and a standard deviation of 0.0303. Primary terms of the sample leases ranged from one to ten years, with a mean term of 3.53 years and a standard deviation of 1.32 years.

The payment structure and other aspects of the lease contracts in the sample must then be correlated with characteristics of nearby producing fields in order to calculate measures of resource value uncertainty and other theoretical constructs. To do so, leases were located on detailed oil and gas production maps using the location description provided for each lease. Distances from each lease location to the nearest productive fields were noted. This distance was taken to be that between the center of the lease in question to the nearest productive well in each of the three nearest fields. Because of the detail of the production maps and the specificity of the reported lease locations, these distances could be recorded with accuracy ranging from one-fifth to one-tenth mile. In general, the sample leases are located relatively close to productive oil and gas fields. Of the 140 sample leases, 88 are located within a mile or less of oil or gas fields.

In addition to the distance to productive fields, the production characteristics of those fields were also recorded. Data on oil and gas fields is available on a subscription basis and includes annual data on the amount of oil, gas, and water produced; the number of active oil wells and gas wells; the total number of wells; and the cumulative production of oil, gas, and water. Other data on field characteristics, including the field discovery dates and shut-in dates (if applicable), were obtained from annual production reviews published by state agencies or oil and gas conservation boards.

Econometric Analysis

To test the predictions of the transaction cost model, the sample contracts are categorized in terms of whether the payment structure emphasizes royalty payments or bonus payments. The effects of the independent variables on the type of payment structure chosen are examined with the use of limited dependent variable techniques.

The predictions forthcoming from the transaction cost model focus on the degree to which contracts emphasize royalty payments relative to bonus payments. The question arises, then, of how to compare contracts with different combinations of royalty rates and bonus payments. This is more difficult in the case of leasing on private land rather than on public land, where law fixes the royalty rate and bidding takes place on the basis of bonus payments alone. (8) Without the constraints on the bidding mechanism that exists on public land, bidding can take place over both the royalty rate and bonus payments simultaneously in the case of leasing on fee land. The fact that this occurs on fee land is not surprising. In fact, Samuelson [1986] demonstrated that a seller can do strictly better utilizing a bidding mechanism, termed mixed bidding, in which neither of the fixed or conditional components are constrained, than when one of these parameters is fixed.

With mixed bidding for oil and gas leasing, the value of total payments with bonus and royalties is determined by a linear payment function:

P = B + [RV.sup.9] (1)

where P is the payment to the seller by the winning bidder, B is the bonus payment, R is the royalty rate, and [V.sup.9] is the ex post discounted value of gross revenues over the production period. If the resource value was known, the present value of royalty payments can be compared with the value of bonuses. Because lease contracts are signed prior to drilling, however, the value of any petroleum reserves associated with the lease is unknown. For the purpose of testing the predictions of the transaction cost and risk-based models, a complete analysis of the mixed bidding mechanism is not necessary. The predictions can be evaluated by examining the determinants of the variation in the royalty relative to the bonus payment component of petroleum leasing contracts. This can be accomplished by holding one of the two components constant and examining changes in the other component.

The method employed here is to aggregate contracts by bonus payments and examine variation in royalty rates. (9) This approach is facilitated by the fact that the sample contracts tend to be clustered into four reasonably distinct bonus payment groups. The clustering of the sample contracts by bonus payments enables the examination of the variation within each bonus group. This is done using a single-stage estimation procedure that looks at variability in royalty rates within each group. This essentially holds bonus payments constant and regresses the royalty rates in each bonus group against the exogenous variables described above. Within each bonus group, variation in the royalty rate will result from variation in the exogenous variables.

Contracts within each bonus group were classified as either share contracts (those with relatively high conditional components) or fixed contracts (those with relatively higher fixed payment components). A newly generated dependent variable, SHARE, takes on a value of 0 if classified as a fixed payment type contract and 1 for contracts classified as a share-type contract. (10) With the construction of the binary dependent variable, the estimating equation becomes:

SHARE = [[alpha].sub.1] + [[beta].sub.1]DIST+ [[beta].sub.2]VAR + [[beta].sub.3]FIELDSIZE + [[beta].sub.4]FIRMSIZE + [[beta].sub.5] PARTIES + [[beta].sub.6]TERM + [epsilon]. (2)

Logit regressions can then be performed on this estimating equation. The parameter estimates, standard errors, t-statistics, and the signs of the coefficients predicted from the transaction cost and risk-sharing model are presented in Table 1.

The results of this logit estimation are generally supportive of the transaction costs model. With the exception of the VAR variable, the coefficients are consistent with the signs predicted by the transaction cost analysis. The coefficients on DIST and FIRMSIZE are positive and significant at the 5 percent level, as predicted by the transaction costs model. This means that there is a greater probability of higher royalty rates with an increase in uncertainty of resource value (as measured by DIST) and an increase in the costliness of bidders undertaking pre-sale measurement (as measured by FIRMSIZE). The coefficient of the PARTIES regressor is negative and is statistically significant. This lends support to the notion that negotiation costs are an important determinant of contract structure in oil and gas leasing. The coefficient on the TERM and FIELDSIZE regressors have signs as predicted by the transaction cost model, but are significant at less than the 10 percent level. Overall, the results indicate that the costs of pre-sale measurement are important influences on contract structure, that negotiation costs may also be an important factor, but that losses from delayed production or non-production due to the presence of high royalties may not significantly influence contract structure.

Summary of Findings and Conclusion

The process of exploring for and developing petroleum resources encompasses a series of operations involving a number and variety of input owners. Early in the process, owners of exploration and drilling inputs contract with owners of potential in-situ oil and gas deposits. The form of these contracts can be complex and varies across countries and institutional settings. This study examines the payment structure of leasing contracts for petroleum exploration and development in the United States. The principal focus of the study is the development of a model of contract choice aimed at examining the determinants of the relative importance of conditional and lump-sum payment provisions in mineral rights leasing contracts for oil and gas development.

The conventional explanation of the payment structure of oil and gas leasing contracts is that the conditional payment provisions are a means to share this risk between the buyer and the seller. The alternative explanation offered by this study focuses on differences in the transaction costs associated with the different payment types. The pattern of variation in the payment structure of the sample leases is found to be generally consistent with the transactions costs model. This is especially true for the predicted relationship between the degree of resource value uncertainty and the conditional component of the total payment structure. As distance of the lease locations from known producing fields increases, there is an increase in the degree to which leasing contracts emphasize royalty payments relative to bonus payments. In the transaction cost framework, this is explained by efforts on the part of landowners to reduce the incentives for potential lessees to engage in costly presale measurement activities . In the transaction cost model, the use of royalties to reduce the effects of pre-sale measurement is tempered by the incentive generated by royalties to underproduce the resource.

The model developed here has implications for the ongoing discussion on revising federal policy toward resource development on federal lands. Recommendations to increase the statutorily fixed royalty rates for oil and gas leasing on OCS tracts and to increase bonus payments for mineral development on onshore public lands often fail to consider the costs of modifying payment provisions involving both fixed and conditional components. The results of this study show, for example, that increasing royalty rates will result in higher losses due to speculation and negotiation and enforcement costs. On the other hand, increasing the fixed payment component will, ceteris paribus, increase losses due to underproduction. Revising federal leasing policy for natural resources in an optimal manner involves a close examination of the all of the transaction costs involved.
TABLE 1

Results of Logit Analysis

            Estimated     Standard                  Predicted
Parameter  Coefficient     Error       t-statistic    Sign

[alpha]      -2.214        1.153         -1.918         *
DIST          0.201        0.088          2.277         +
VAR          -0.889 E-03   0.142 E-02    -0.628         +
FIELDSIZE     0.957 E-03   0.137 E-02     0.695         +
FIRMSIZE      1.155        0.301          2.948         +
PARTIES      -1.809        0.742         -2.437         -
TERM         -0.297        0.240         -1.238         -

Notes: * No predicted sign.

[R.sup.2] = 0.47034, percentage of correct predictions is 84.29 percent.


Footnotes

(1.) Commencing a producing well will extend the lease for as long as production continues.

(2.) In some cases, annual rental payments are also made. Like bonuses, annual rentals are a fixed payment per acre leased. Unlike bonuses, which are paid once when the contract is signed, rentals are paid annually for the initial term of the lease. Relatively few contracts contain annual rentals and, for the purposes of the current discussion, annual rentals are considered as part of bonus payments.

(3.) This is in contrast to second-price auctions in which the second highest bid wins, and in contrast to open, or English, auctions, in which bidders are able to observe their rivals' bids.

(4.) A third possibility, that of informational rents being driven to zero by having an infinite number of bidders, is rejected.

(5.) The model by French and McCormick [1984] synthesizes elements from models including those by Wilson [1977], Gaskins and Teisberg [1979], Johnson [1979], and Ramsey [1980].

(6.) See Gaskins and Teisberg [1979], Englebracht- Wiggans, Milgrom, and Weber [1983], and Milgrom and Weber [1982a, 1982b].

(7.) Due to production decline over time, a large field near the end of its production horizon may yield the same measure of annual production as a small field recently brought under production. Similarly, with cumulative production as a measure, a small field under production for a long period of time may yield the same measure as a large field recently discovered.

(8.) The exception to this is the experimental program of OCS bidding in the late 1970s, in which the bonus payment was fixed and bidding took place on the basis of the royalty rate.

(9.) An alternative is to hold royalty payments constant and examine changes in bonus payments. This method is implicit in studies using federal leasing data because of the use of fixed-royalty bidding by the federal government. For mixed bidding, this approach involves examining variation in bonus payments among contracts with the same royalty rate. The preferred method for doing this is to use a two-stage least squares approach in which the bonus payment variable is regressed against the relevant independent variables in order to get the predicted values of the bonus payment variable. Then, in the second stage, the predicted values of bonus payments would be used as instruments along with the independent variables. Unfortunately, because the two-equation simultaneous equation system has the same exogenous variables in each of the structural equations, this approach fails the identification test.

(10.) Contracts with royalty rates higher than the group mean were assigned a SHARE value of 1. Those with royalty rates below the mean for the group were assigned a value of 0 for SHARE.

References

Allen, Douglas; Lueck, Dean. "The Role of Risk in Contract Choice," Journal of Law, Economics, and Organization, 15, 3,1999, pp. 704-36.

Barzel, Yoram. "Measurement Cost and the Organization of Markets," Journal of Law and Economics, 25, 1, 1982, pp. 27-48.

-----. "The Entrepreneurs Reward for Self-Policing," Economic Inquiry, 25, 1, 1987, pp. 103-16.

Black, Geoffrey. "The Incentive Effect in Share Contracts: The Case of Finite Resources," International Advances in Economic Research, 6, 3, 2000, pp. 461-74.

Cheung, Steven N. S. "Transaction Costs, Risk Aversion and the Choice of Contractual Arrangements," Journal of Law and Economics, 12, 1969, pp. 23-42.

Dam, Kenneth W. Oil Resources: Who Gets What How?, Chicago: University of Chicago Press, 1976.

Engelbracht-Wiggans, Richard; Milgrom, Paul; Weber, Robert. "Competitive Bidding and Proprietary Information," Journal of Mathematical Economics, 11, 1983, pp. 161-9.

French, Kenneth; McCormick, Robert. "Sealed Bids, Sunk Costs, and the Process of Competition," Journal of Business, 57, 4, pp. 417-41.

Gaskins, Darius; Teisberg, Thomas. "An Economic Analysis of Presale Exploration in Oil and Gas Lease Sales," In Masson, Robert and P. David Qualls, eds., Essays on Industrial Organization in Honor of Joe Bain, Cambridge: Ballinger Publishing Co., 1979.

Grossman, Sanford J.; Hart, Oliver D. "An Analysis of the Principal Agent Problem," Econometrica, 51, 1,: 1983, pp. 7-45.

Hendricks, Kenneth; Porter, Robert. "An Empirical Study of an Auction with Asymmetric Information," American Economic Review, 78, 1988, pp. 865-83.

Hendricks, Kenneth; Porter, Robert; Tan, Guofu. "Optimal Selling Strategies with an Informed Buyer," American Economic Review: Papers and Proceedings, 83, 1993, pp. 234-9.

Hendricks, Kenneth; Porter, Robert; Wilson, Charles. "Auctions for Oil and Gas Leases with an Informed Bidder and a Random Reservation Price," Econometrica, 62, 1994, pp. 1415-44.

Johnson, Ronald N. "Auction Markets, Bid Preparation Costs and Entrance Fees," Land Economics, 55, 3,1979, pp. 313-8.

Leffler, Keith; Rucker, Randal. "Transaction Costs and the Efficient Organization of Production: A Study of Timber-Harvesting Contracts," Journal of Political Economy, 99, 5, 1991, pp. 1060-87.

Leland, Hayne E. "Optimal Risk Sharing and the Leasing of Natural Resources, with an Application to Oil and Gas Leasing on the OCS," The Quarterly Journal of Economics, 92, 3, 1978, pp. 413-37.

Leland, Hayne E.; Norgaard, Richard; Pearson, Stephen. "An Economic Analysis of Alternative Outer Continental Shelf Petroleum Leasing Policies," Office of Energy R and D Policy Report, National Science Foundation, 1974.

McAfee, Preston; McMillan, John. "Bidding for Contracts: A Principal-Agent Analysis," Rand Journal of Economics, 17, 1986, pp. 326-38.

Mead, Walter; Moseidjord, Asbjorn; Muraoka, Dennis; Sorensen, Philip. Offshore Lands, San Francisco: Pacific Institute for Public Policy Research, 1985.

Milgrom, Paul; Weber, Robert. "The Value of Information in a Sealed-Bid Auction," Journal of Mathematical Economics, 10, 1982, pp. 105-14.

Ramsey, James B. Bidding and Oil Leases, Greenwich: JAI Press, 1980.

Reece, D. K. "Competitive Bidding for Offshore Petroleum Resources," Bell Journal of Economics, 9, 2, 1978, pp. 369-84.

-----. "An Analysis of Alternative Bidding Systems for Leasing Offshore Oil," Bell Journal of Economics, 8, 2, 1979, pp. 659-69.

Rucker, Randal; Leffler, Keith. "To Harvest or Not to Harvest? An Analysis of Cutting Behavior on Federal Timber Sales Contracts," Review of Economics and Statistics, 70, 1988, pp. 207-13.

Samuelson, William. "Bidding for Contracts," Management Science, 32, 2, December 1986, pp. 15 33-50.

Williamson, Oliver E. "Transaction-Cost Economics: The Governance of Contractual Relations," Journal of Law and Economics, 22, 1979, pp. 233-61.

Wilson, Robert. "A Bidding Model of Perfect Competition," Review of Economic Studies, 44, 3, 1977, pp. 511-8.

GEOFFREY BLACK *

* Boise State University--U.S.A.
COPYRIGHT 2002 Atlantic Economic Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2002 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Comment:A transaction cost model of contract choice: The case of petroleum exploration.
Author:Black, Geoffrey
Publication:International Advances in Economic Research
Article Type:Brief Article
Geographic Code:1USA
Date:Aug 1, 2002
Words:7355
Previous Article:Market revaluations of foreign listings' reconciliations to U.S. financial reporting.
Next Article:The effect of task experience on assessments of auditor expert potential.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters