volume 58 article #3

Financial Structure and Efficiency of Grain Farms

Raoul E. Nasr, Peter J. Barry, and Paul N. Ellinger

Nasr is an assistant professor of agricultural economics, Jordan University of Science and Technology; Barry is a professor and Ellinger an associate professor, both in the Department of Agricultural and Consumer Economics, University of Illinois.

Abstract

The relationships between farm-level efficiency and financial structure are evaluated using agency cost, free cash flow, and credit evaluation concepts of finance theory. A nonparametric analysis of efficiency is performed based on a sample of 154 Illinois farmers over a seven-year period from 1988­94. The results suggest a positive relationship between efficiency and financial structure, where the linkage is reflective of the motivation provided by Jensen’s free cash flow concept and the credit evaluation concept. These findings are further evidence of a lack of separation between financing and production, as is implied by financial theory.

Key words: efficiency, nonparametric, financial structure, agency costs, free cash flow.

Article <top>

The degree of technical efficiency attained by farm businesses contributes directly to resource productivity and provides a clear signal of managerial effectiveness. Thus it is a significant determinant of business performance. Efficiency has been studied extensively in both developed and developing economy settings (e.g., Bravo-Ureta and Rieger; Bravo-Ureta and Pinheiro 1993, 1997; Tauer; Färe, Grabowski, and Grosskopf; Dawson, Lingard, and Wood). Substantial insight has been gained on the relationship between efficiency and various business and demographic characteristics.

Only a few studies, however, have sought to consider the relationships between efficiency and a farm’s financial arrangements. Employing data envelopment analysis, Whittaker and Morehart found that one in five midwestern cash grain farms is unable to operate at the best-practice, cost-efficient frontier due to debt and/or asset value constraints. Lee and Chambers, and Färe, Grosskopf, and Lovell also found evidence of expenditure-constrained profit maximization in agriculture. Similarly, Chavas and Aliber observed significant linkages between the efficiency of Wisconsin farms and their intermediate and fixed financial structures, but not for current financial structure.

The agency cost, free cash flow, and credit evaluation concepts of finance theory provide alternative explanations for the potential relationship between financing arrangements and farm-level efficiency. None of the studies cited above, however, considered the relationships between efficiency and financial structure within the testable, yet conflicting rationales provided by these alternative concepts from finance. The goal of this study is to analyze empirically the relationships between farm-level efficiency and financial structure using the agency cost, free cash flow, and credit evaluation concepts of finance theory.

The results of this analysis are important, because they will provide empirical insight about the relationships between efficiency and financial structure. The potential relationships between the finance concepts and efficiency are discussed in the following section. The empirical analysis then proceeds by measuring the technical efficiencies of a panel of Illinois grain farmers and by testing the relationships between the efficiency measures and a set of efficiency determinants including financial structure, farm size, tenure position, soil productivity, time, and age of farm operator.

Financial Concepts and Farm-Level Efficiency <top>

The agency cost concept in finance, originated by Jensen and Meckling, hypothesizes that monitoring, bonding, and adverse incentive costs are incurred in a borrower-lender relationship in order to resolve problems of asymmetric information and misaligned incentives between the two parties. Monitoring involves the lender’s observation of the borrower’s progress and activities in order to gain influential information about the prospects of loan repayment. Bonding involves the pledging by borrowers of collateral, reputation, co-signees of notes, and other indicators to persuade lenders that the borrowers are acceptable credit risks. Adverse incentives reflect self-interest-seeking actions by borrowers (excessive risk taking, unplanned capital expenditures, unintended use of borrowed funds) that hamper loan performance and constrain the complete alignment of the borrower’s incentives with those of a lender. Agency costs are largely passed on by lenders to borrowers through adjustments in interest rates, origination fees, servicing fees, collateral requirements, and so on (Ellinger and Barry; Miller, Ellinger, Barry, and Lajili). These costs, in turn, may detract from the borrower’s technical efficiency and impede business performance.

In agriculture, the agency cost concept also can apply to leasing contracts between land owners (the principal) and farm operators (the agent). Agency costs for the land owner may arise because information is lacking about the farmer’s characteristics when the contract is initiated (a potential adverse selection problem resulting in a low-quality farmer) and about the farmer’s propensity for altering anticipated plans during the course of the lease contract (potential moral hazards such as shirking, deception, or other opportunistic, self-interested behavior). These problems reflect the combined effects of asymmetric information and misaligned incentives between the contracting parties. Various practices may be utilized to reduce these costs. Included are share rents to achieve incentive alignments between the two parties, monitoring and reporting requirements, collateral pledges by the operator, and reputation effects in the local leasing market.

The free cash flow concept, developed by Jensen to explain debt-financed corporate takeovers, suggests that agents with excess cash flows and abundant financial assets may exercise managerial laxness, devote insufficient attention to detail, and squander resources in nonbusiness uses. The general effects are a diminution in the firm’s financial performance and increased vulnerability to mergers, acquisition, or other losses of business independence. A possible solution to these maladies is the creation of leverage-induced financial obligations that will stimulate increased effort by agents to satisfy these obligations. In the agricultural setting, the concept suggests greater effort by farmers to meet greater debt obligations. The increased effort will enhance technical efficiency, and thus improve business performance unless the obligations themselves become excessive.

Only a few empirical tests of the free cash flow theory have been conducted, all in corporate finance. Kim and Maksimovic considered the relationships between agency cost and free cash flow concepts to efficiency in the airline industry and found evidence supporting the agency cost concept. Lehn and Poulsen reported a significant positive relationship between free cash flow and a firm’s decision to go private. They also found that premiums paid to stockholders are positively related to free cash flows. Kallapur confirmed positive support for the free cash flow theory by testing for a positive relationship between earnings response coefficients and dividend payout ratios for a sample of 112 firms over the 1951­86 period. Using a sample of successful tender offers, Lang, Stulz, and Walkling used Tobin’s q and several cash flow measures to show that bidder returns are significantly negatively related to cash flow for low q bidders, but not for high q bidders.

The credit evaluation concept suggests that lenders will prefer to finance more efficient farmers because these borrowers have lower credit risks. Evidence about the composition of credit scoring models (Ellinger, Splett, and Barry), for example, indicates that agricultural bankers often use management/efficiency variables (i.e., operating costs per acre, yield per acre, profit per cow, etc.) along with various financial variables in evaluating creditworthiness. Use of greater financial leverage by some borrowers could represent certification of their greater technical efficiency, through the lender’s favorable evaluation of the borrower’s creditworthiness and the resulting loan decisions. The comparative advantages of banks and other local lenders in collecting, analyzing, and monitoring information about borrowers means that the existence of loans from these lenders certifies their anticipations for satisfactory loan performance (Fama; James and Weir).

The credit evaluation concept is consistent with the liquidity preference theory of credit use in agriculture, developed by Baker and extended by Baker and Hopkin and by Barry, Baker, and Sanint. Liquidity preference theory suggests that lenders’ preferences for asset-generating, self-liquidating loans result in greater credit availability, lower interest rates and/or other favorable financing terms for these preferred loan purposes. Greater technical efficiency achieved by borrowers is consistent with the liquidity preference theory because it will augment repayment prospects and add to creditworthiness, unless the enhanced efficiency skills compete adversely with the farmer’s skills in marketing, investment analysis, financial management, and other areas.

In summary, the agency cost concept implies a negative relationship between technical efficiency and financial structure, while the free cash flow and credit evaluation concepts suggest positive relationships. Similarly, the agency cost concept applied to leasing would be supported by a reduction in agency costs between the tenant and the land owner as the proportion of land owned (tenure) by the operator increases. This study will test for a positive relationship of efficiency with the free cash flow and credit evaluation concepts, and for a negative relationship with agency costs. Moreover, the free cash flow and credit evaluation effects will be distinguished through the form of the financial structure variable.

The form of the financial structure variable involves the distinction between operating credit (to finance production and marketing) and capital credit (to finance durable assets). Barry, Baker, and Sanint found that agricultural lenders constrain capital credit more than operating credit in response to variations in recent financial performance due to factors beyond the farmers’ control.

Availability of operating credit is therefore more stable than capital credit under exogenous risk conditions. It is plausible, however, that operating credit will be more responsive to performance variations that are controllable by farmers, including the degrees of effort devoted to efficiency improvement. Moreover, operating (or current) debt used to finance production, inventories, current payments of fixed debt, and other short-term obligations is the nearest-in-time financial obligation faced by farmers. It likely comes from lenders with whom farmers have the closest customer relationships, and thus provides the best opportunity for the lenders to observe the farmers’ individual efforts and performance.

It is logical, therefore, to expect that operating credit is more strongly related than capital credit or total credit to farmers’ efficiency in terms of either agency costs or farmers’ efforts to meet these financial obligations. For longer term performance relative to the total of operating debt plus capital debt obligations, technical efficiency is combined more fully with other determinants of creditworthiness.

To test the free cash flow concept, it would be ideal to identify a variable that discriminates well between high and low cash flows. However, because detailed cash flow data are unavailable for the entire data set used in this study, the ratio of current debt-to-total assets is used as a proxy. Current debt includes operating loans, current payments on term debt, and other short-term obligations. A significant positive relationship between the current debt-to-total asset ratio and technical efficiency will be consistent with the free cash flow concept, whereas a significant positive relationship between total debt and total assets will be more consistent with the credit evaluation concept. For the latter concept, confirming a positive relationship would support the findings of Ellinger, Splett, and Barry—i.e., that the total debt-to-asset ratio receives the greatest weight by lenders in credit evaluation.

Data Source <top>

The data for this analysis are obtained from Illinois grain farms that maintain production and financial records with the Illinois Farm Business Farm Management Association (IFBFMA). This data source has been used extensively for research and extension purposes over a long period of time. A panel of 154 grain farmers who maintained complete production and financial records for the continuous seven-year period from 1988 through 1994 is used in the analysis.

The theoretical framework of efficiency measurement is based on the use of quantity data for inputs and outputs. The IFBFMA data are primarily accounting data. Quantity data such as pounds of chemicals, number of seeds used, and hours of labor are unavailable. In the absence of such quantity data, expenditure and revenue data are often used as proxies for input and output quantities (Aly, Belbase, Grabowski, and Kraft; Grabowski, Kraft, Pasurka, and Aly; Neff, Garcia, and Hornbaker).

Nearly all of the farm revenue for the 154 farms is derived from crop production. The average farm size in 1994 was 846 acres with an average gross revenue of $244 per acre₁. Total costs averaged $196 per acre, including accrual-based expenditures on fertilizer, chemicals, seeds, power and equipment, paid and unpaid labor, depreciation, and other expenses (drying and storage, supplies and services, building repairs, insurance, taxes, and miscellaneous). Land costs were excluded from the analysis since utilizing the same opportunity cost rates of land for each farm would have resulted in uniform effects across farms with no differential effects on the efficiency measures. Additional details about the levels of the revenue and expenditure data are provided in Table 1.

¹All revenue and costs are reported on an operator basis.

Table 1. Revenue and Expenditure Data for the Sample of 154 Illinois Grain Farms, 1994 ($/acre)

 

Model Specifications 

Technical Efficiency <top>

Technical efficiency of farmers can be measured using parametric or nonparametric frontier techniques. In this study, Farrell’s nonparametric, input-based measure of technical efficiency is computed and decomposed using mathematical programming into its three components: (a) scale efficiency, (b) pure technical efficiency, and (c) a congestion-free measure (Färe, Grosskopf, and Lovell).  Each measure is computed assuming one output (gross farm revenue) and six inputs (chemicals and fertilizer, seed, power and equipment, labor, depreciable capital, and other input expenses). Thus, the overall technical efficiency measure and each component are computed for each farm to obtain the degree of inefficiency arising from (a) failure to operate at optimal size, (b) pure technical inefficiency, and (c) congestion arising from the over-utilization of some inputs.

Data envelopment analysis (DEA) is a well-accepted and widely used approach to measure firm-level efficiencies. Seiford provides a discussion of the evolution of DEA and reports a bibliography of over 700 studies using the DEA approach for various applications. The efficiency measures are straightforward to estimate empirically. The nonparametric approach to efficiency measurement does not impose functional restrictions on technology, as typically occurs using an econometric approach to efficiency measurement. It is based on the solutions of appropriately formulated mathematical programming models and can easily handle disaggregated inputs. Another advantage is that DEA provides farm-specific information on the components of technical efficiency.

A drawback of the nonparametric approach is the implication that all deviations from the frontier are interpreted as inefficiencies. Other problems with the approach are the lack of statistical inference associated with the efficiency estimates and the implication that an inefficient farm can only be brought back toward the efficient frontier by shrinking all inputs in the same proportion.

As explained by Färe, Grosskopf, and Lovell, the Farrell input-based measure of technical efficiency is computed using a constructed reference technology that satisfies strong disposability of inputs (SDI) and constant returns to scale (CRS).  Formally, these restrictions are: L(y) admits strong disposability of inputs if g Ž d ε L(y), g ε L(y); and L(y) satisfies constant returns to scale if L(θy) = θL(y), for θ Ž 0, where L(y) is the input requirement set for output level y (see Figure 1). The piecewise linear technology constructed from the observed inputs and outputs (x, y), and satisfying SDI and CRS, can be written as:

    (1)

For a given observation of the input and output vectors, denoted x°, y°, the Farrell input-based technical efficiency is measured by:

   (2)

The measure is computed by solving the following linear programming problem:  

   (3)

The elements of this linear programming problem and the efficiency measures are illustrated using the three observations, a, b, and g in Figure 1. The technology that envelopes these data is the lower bound of L^k(y), labeled y° in Figure 1. Observations a and b are technically efficient, operating on and not inside the bound of L^k(y); hence, E_i(x°, y°) equals 1.0 for these observations. Only observation g is technically inefficient relative to this reference technology. For observation g, E_i(x°, y°) < 1, and it is measured by the radial distance (od/og) from the observed point to the reference technology. The level that inputs could be proportionally reduced without changing output level y° is given by 1 ­ E_i(x°, y°).

Färe, Grosskopf, and Lovell’s method for identifying the components of inefficiency is also employed. The method involves relaxing the assumptions that the constructed reference technology satisfies SDI and CRS. The result is a decomposition of the overall technical efficiency measure E_i(x°, y°) into the three mutually exclusive and exhaustive components of: scale efficiency S_i(x°, y°), pure technical efficiency F_i(x°, y°), and input congestion C_i(x°, y°), so that

(4)

These measures are computed as solutions to the linear programming problems, using GAMS software, with constraints that describe the less restrictive, piecewise linear representations of the technology.

Although the data set created for this study is a panel, nonparametric efficiency measures are computed for each year independently. A shortcoming of the nonparametric approach is its limited capability to utilize panel data in a straightforward manner. Kneip and Simar provide the first study to develop a technique for estimating frontiers with the nonparametric approach using panel data. Their technique is complex and assumes efficiencies are independent over time. Moreover, little empirical work has been done to evaluate the robustness of their method. The parametric approach to panel data has been developed and evaluated (e.g., Cornwall, Schmidt, and Sickles; Battese and Coelli; Kumbhakar); however, a specific time variance efficiency structure had to be imposed in each of these studies.

Business and Demographic Characteristics <top>

While this study focuses on the relationship between a farm’s efficiency and its financial structure, it is important to account for the potential effects of other factors on efficiency. A positive relationship between technical efficiency and farm size is expected because of economies of scale. Smaller farms tend to be less efficient and lie farther away from the efficient frontier (Hall and Leveen). A positive relationship between size and efficiency was observed by Aly et al. for Illinois grain farms, by Bagi and Huang for Tennessee grain and mixed farms, by Grisley and Masccarenhas for Pennsylvania dairy farms, and by Tauer and Belbase for New York dairy farms. In contrast, Bravo-Ureta and Rieger found that the efficiency of New England dairy farms is not markedly affected by farm size, and Garcia, Offutt, and Sonka concluded that small Illinois grain farms are just as efficient as larger farms.

The potential effect of tenure (acreage owned relative to acres owned plus acres leased) on efficiency is ambiguous. A negative relationship would be consistent with the Jensen free cash flow concept in that greater leasing may compel farmers to work harder to meet their contractual obligations with multiple landlords and signal their lease-worthiness. Alternatively, a positive relationship would be consistent with agency models that predict a decrease in efficiency as leasing increases (Braverman and Stiglitz), perhaps reflecting monitoring problems and adverse incentives between landlords and tenants that hamper production plans and diminish business performance.

As a farmer ages and gains experience, he or she may realize improved management abilities that contribute to greater efficiency. Tauer, for example, used cross-sectional data from the 1978 Agricultural Census to conclude that farmers display increases and then decreases in productivity over their life cycles, with the greatest productivity in the 35­44 year age range.

Farmers with higher quality soil also may exhibit higher levels of technical efficiency, perhaps reflecting the use of different types of technology, and differences in soil topography, uniformity, drainage, and other quality attributes.

For our analysis, the efficiency measures are estimated for seven years (1988­94) with a separate frontier for each year. Specific annual changes in factors like weather, insects, and commodity price levels may affect the distribution of the efficiency measures from year to year. Thus, in the combined panel data set, dummy variables for each year are evaluated.

Two-Step Empirical Model <top>

A commonly used two-step procedure approach is employed to explain variation of efficiency scores across firms (Grosskopf; Bravo-Ureta and Pinheiro 1997). The first step is to estimate the efficiency scores, and the second step is to explain the variation of the efficiency scores using appropriate regression procedures.

A few conceptual issues arise with this approach. First, the efficiency scores (scale efficiency, pure technical efficiency, and congestion-free efficiency) have an upper bound of 1.0 and a lower bound of 0.0₂. Thus, ordinary least squares estimates are inconsistent (Greene). A tobit regression model is one way to overcome these problems (Greene; Chavas and Aliber; Bravo-Ureta and Pinheiro 1997). The general formulation of the tobit model is given in terms of an index function:

   (5)

where ES is one of the efficiency measures.

²The proportion of observations for overall technical efficiency, scale efficiency, pure technical efficiency, and congestion-free efficiency that are bounded at 1.0 are 19%, 19%, 68%, and 38%, respectively. There are no observations with 0% efficiency. The empirical results do not change for an upper-limit-only tobit model.

Another issue with the two-step approach arises from the data-generating process. When the variables used to define the data in step one are highly correlated with the variables used in the second step, the estimates from the second step will be inconsistent (Deprins and Simar). With the exception of the correlation of labor expense and acres (-0.54), the absolute values of all correlation coefficients from the inputs in step one and the independent variables in step two are below 0.19. Thus, the inconsistency issue is not a large concern with the two-step approach applied in this study.

Another assumption of the two-step procedure is that the independent variables in step two affect efficiency. If this is the case, then conceptually, the variables should have been included as input variables in the estimation of the first step. However, statistical tests of significance of the input variables in the first step are difficult to obtain using the nonparametric procedure.

The second step of the two-step model in this study is expressed as:

(6)

where Y_i is the measure of technical efficiency, x₁ is financial structure, x₂ is farm size, x₃ is tenure position, x₄ is the operator’s age, x₅ is soil productivity, and x₆ is the set of year dummies. The different measures of technical efficiency are total technical efficiency (Y₁), scale efficiency (Y₂), pure technical efficiency (Y₃), and congestion-free efficiency (Y₄).

The pure technical efficiency measure is more appropriately related to the agency cost, free cash flow concepts, and credit evaluation concepts. These concepts relate to the ability of the firm to operate without wasting inputs rather than operating at an inappropriate scale. The scale efficiency model is also suited for the credit evaluation concept since lenders consider the appropriate scale of the operation in addition to financial characteristics and pure technical efficiency measures.

Efficiency Measures <top>

The efficiency measurements estimated for the 1988­94 period under constant returns to scale conditions are shown in Table 2. In 1994, for example, the average technical efficiency, assuming constant returns to scale and strong disposability of inputs, is 86.9%. This result suggests that expenses could have been reduced by an average of 13.1% without affecting total output. As shown in Table 2, the averages of the efficiency components for 1994 are: scale efficiency 93.2%, pure technical efficiency 96.6%, and congestion-free efficiency 96.6%. The value of overall technical efficiency (Y₁) must be less than or equal to its components since Y₁ is the product of scale, pure technical, and congestion-free efficiency.

Over the 1988­94 period, average technical efficiency ranged from 76.5% in 1988 to 86.9% in 1994 (Table 2). The averages of the efficiency components ranged from 94% to 96.6% for pure technical efficiency, from 87.8% to 93.2% for scale efficiency, and from 92.8% to 96.9% for congestion-free efficiency. In 1994, the frequency distribution for the 154 farms shows that 40.3% have an overall technical efficiency of less than 85%, 28.6% have a rating from 85­95%, 6.5% are rated from 95­99%, and 24.7% are rated from 99­100%, inclusively.

In order to gain insight on some of the relationships among the variables, the farms are classified by efficiency level. Table 3 shows a comparison of mean values of business characteristics across low, medium, and high efficiency categories, where the boundaries between the medium class and the low and high classes reflect the lower and upper quartiles of the overall efficiency distribution. The respective mean values for 1994 indicate that measures of farm acreage, total assets, debt-to-asset ratio, and the ratio of current debt-to-total assets tend to increase as the efficiency class increases. No clear patterns are evident for the other variables. To measure the possible multivariate relationships more explicitly, a tobit regression is performed.

 

 

Regression Analysis <top>

The results of the tobit model for overall technical efficiency and its components are reported in Table 4. Two sets of regression equations are run for the alternative specifications of the financial structure variable because the debt-to-asset ratio and current debt-to-asset ratio tend to be positively correlated. Thus, one set of regressions includes the ratio of current debt-to-total assets along with the other independent variables, while a second set of regressions includes the debt-to-asset ratio together with the other independent variables.

 

 

No significant relationships are found between the efficiency measures and the debt-to-asset ratio₃. Furthermore, the coefficients on age variables are insignificant in all equations, and thus excluded from the regression analysis. The equations are reestimated with tenure, acres, soil rating, the current debt-to-asset ratio, and the year dummies as the independent variables₄. The regression results are reported in Table 4.

³The squares of the debt-to-asset ratio and the current debt-to-total asset ratio also are tested. The square specifications are intended to test whether significance might hold over a limited range of the financial structure variables. All of the coefficients on the square specifications are insignificant and so are not included in the model results reported here.

⁴Additional regressions explored the effects of measuring the financial structure variables by the ratios of current assets-to-current liabilities for operating credit, and intermediate assets-to-intermediate debt and intermediate debt-to-total assets for capital credit. None of these regressions yielded significant results.

 

The log-likelihood measures (Table 4) indicate that the scale efficiency model has the best fit, while the pure technical efficiency model has the weakest fit. The measure of overall technical efficiency is positively and significantly related to the ratio of current debt-to-total assets, the tenure ratio, farm size, and the soil rating. Significant relationships between the component measures of efficiency and the current debt ratio, the tenure ratio, and soil rating occur in two of the three component models. Similarly, farm size is significant in two of the three component models. The effects of heteroskedasticity, autocorrelation, and multicollinearity are minor (Nasr).

The significant positive relationship between technical efficiency and current debt-to-asset ratio provides the strongest support for Jensen’s free cash flow concept. That is, greater reliance by farmers on current debt to finance their operations is consistent with the hypothesis that they will work harder to meet their financial obligations. This same statistical relationship provides some support for the credit evaluation concept, but the lack of a significant relationship between the debt-to-asset ratio and technical efficiency fails to substantiate the credit evaluation concept.  

Apparently, a broader set of management and financial factors, beyond technical efficiency alone, is associated with evaluations of farmers’ creditworthiness, as suggested by the results of various credit scoring studies (Splett, Barry, Dixon, and Ellinger; Novak and LaDue). Thus, working harder in production to meet debt obligations has a positive, although minor, effect on creditworthiness relative to other factors. While the data fail to support the agency cost concept related to debt financing and efficiency, the significant positive relationship between tenure and technical efficiency supports the agency cost concept between land owners and farm operators.

Concluding Comments <top>

The quantitative results of this study provide promising evidence of a positive linkage between the technical efficiency of farm businesses and their financial structure, where the linkage is reflective of the motivation provided by Jensen’s free cash flow concept, and perhaps the credit evaluation concept as well. That is, farmers with greater financial leverage put forth more productive efforts to meet their debt obligations, although lenders also may be certifying greater creditworthiness associated with more efficient production.

Moreover, these relationships are evident after accounting for the significant effects on technical efficiency of farm size, tenure, and soil productivity.

In contrast, the farmer’s age and overall financial leverage are not found to be significant factors. The positive relationship between tenure and efficiency provides support for the agency cost concept between landlords and tenants. Adverse incentives and monitoring problems between landlords and tenants may influence the tenant’s production practices and operating performance.

The major theoretical implication of these findings is further evidence of a lack of separation between financing and production, as basic finance theory tends to imply. Empirically, the results suggest that lenders may expect their more highly leveraged farm borrowers to work harder to meet their financial obligations, although the limits of such efforts are difficult to determine. It is also likely that higher leverage by farmers certifies greater creditworthiness due in part to higher levels of technical efficiency. In either case, the empirical results suggest that a financial structure variable is appropriate to employ within farm-level efficiency analyses.

The next steps in this line of research could involve the impacts of various lease arrangements on the agency costs between land owners and farmers. Another extension could involve a joint, intertemporal treatment of the interrelationships among financial structure, efficiency, and investment behavior, consistent with the observations by Barry, Baker, and Sanint related to lenders’ differential responses for operating credit and capital credit as net farm income varies over time.  

References <top>

Aly, H.Y., K. Belbase, R. Grabowski, and S. Kraft. "The Technical Efficiency of Illinois Grain Farms: An Application of a Ray Homothetic Production Function." S. J. Agr. Econ. 19,1(1987):69­78.

Bagi, F.S., and C.J. Huang. "Estimating Production Technical Efficiency for Individual Farms in Tennessee." Can. J. Agr. Econ. 31(1983):249­56.

Baker, C.B. "Credit in the Production Organization of the Firm." Amer. J. Agr. Econ. 50(August 1968):507­21.

Baker, C.B., and J. Hopkin. "Concepts of Finance Capital for a Capital-Using Agriculture." Amer. J. Agr. Econ. 51(1969):1055­65.

Barry, P.J., C.B. Baker, and L.R. Sanint. "Farmers’ Credit Risks and Liquidity Management." Amer. J. Agr. Econ. 63(1981):216­27.

Battese, G.E., and T.J. Coelli. "Frontier Production Functions, Technical Efficiency, and Panel Data: With Application to Paddy Farmers in India." J. Productivity Anal. 3(1992):153­69.

Braverman, A., and J.E. Stiglitz. "Cost-Sharing Arrangements Under Sharecropping: Moral Hazard, Incentive Flexibility, and Risk." Amer. J. Agr. Econ. 68(1986):642­52.

Bravo-Ureta, B.E., and A.E. Pinheiro. "Efficiency Analysis of Developing Countries’ Agriculture: A Review of the Frontier Literature." Agr. and Resour. Econ. Rev. 22(April 1993):88­101.

———. "Technical, Economic, and Allocative Efficiency in Peasant Farming: Evidence from the Dominican Republic." Developing Economies 35(March 1997):48­67.

Bravo-Ureta, B.E., and L. Rieger. "Dairy Farm Efficiency Measurement Using Stochastic Frontiers and Neoclassical Duality." Amer. J. Agr. Econ. 73(1991):421­28.

Chavas, J.-P. and M. Aliber. "An Analysis of Economic Efficiency in Agriculture: A Nonparametric Approach." J. Agr. and Resour. Econ. 18,1(1993):1­16.

Cornwall, C., P. Schmidt, and R.C. Sickles. "Production Frontiers with Cross-Sectional and Time-Series Variation in Efficiency Levels." J. Econometrics 46(1990):185­200.

Dawson, P.J., J. Lingard, and C.H. Wood. "A Generalized Measure of Farm-Specific Technical Efficiency." Amer. J. Agr. Econ. 73(1991):1098­1104.

Deprins, D., and L. Simar. "Estimating Technical Inefficiencies with Corrections for Environmental Conditions with an Application to Railway Companies." Annals Public and Coop. Econ. 60(1989): 81­102.

Ellinger, P.N., and P.J. Barry. "Agricultural Credit Delivery Costs at Commercial Banks." Agr. Fin. Rev. 51(1991)64­78.

Ellinger, P.N., N.S. Splett, and P.J. Barry. "Consistency of Credit Evaluation Procedures at Agricultural Banks." Agribus.: An Internat. J. 8(1992): 517­36.

Fama, E. "What’s Different About Banks?" J. Monetary Econ. 15(1985):29­39.

Färe, R., R. Grabowski, and S. Grosskopf. "Technical Efficiency of Philippine Agriculture." Appl. Econ. 17(1985): 205­14.

Färe, R., S. Grosskopf, and C.A.K. Lovell. Production Frontiers. New York: Cambridge University Press, 1994.

Farrell, M.J. "The Measurement of Productivity Efficiency." J. Royal Statis. Soc. 120(1957):253­81.

Garcia, P., S.E. Offutt, and S.T. Sonka. "Size Distribution and Growth in a Sample of Illinois Cash Grain Farms." Amer. J. Agr. Econ. 69(1987): 471­76.

Grabowski, R., S. Kraft, C. Pasurka, and H.Y. Aly. "A Ray-Homothetic Frontier and Efficiency: Grain Farms in Southern Illinois." European J. Agr. Econ. 17(1990):435­48.

Greene, W.H. Econometric Analysis. New York: Macmillan Publishing Co., 1990.

Grisley, W., and J. Masccarenhas. "Operating Cost Efficiency on Pennsylvania Dairy Farms." Northwest. J. Agr. and Resour. Econ. 14(1985): 88­95.

Grosskopf, S. "Statistical Inference and Nonparametric Efficiency." J. Productivity Anal. 7(July 1996): 161­76.

Hall, B., and F.P. Leveen. "Farm Size and Economic Efficiency: The Case of California." Amer. J. Agr. Econ. 60(1978):589­600.

James, C., and P. Weir. "Borrowing Relationships, Intermediation, and the Cost of Issuing Public Securities." J. Finan. Econ. 28(1990):149­71.

Jensen, M.C. "Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers." Amer. Econ. Rev. 76(May 1986):233­38.

Jensen, M.C., and W.H. Meckling. "Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure." J. Finan. Econ. 3(October 1976):305­60.

Kallapur, S. "Dividend Payout Ratios as Determinants of Earnings Response Coefficients." J. Accounting and Econ. 17(1994):359­75.

Kim, M., and V. Maksimovic. "Debt and Input Misallocation." J. Finance 45(July 1990):795­816.

Kneip, A., and L. Simar. "A General Framework for Frontier Estimation with Panel Data." J. Productivity Anal. 7(July 1996):187­212.

Kumbhakar, S.C. "Production Frontiers, Panel Data, and Time-Varying Technical Inefficiency." J. Econometrics 46(1990): 201­11.

Lang, L., R. Stulz, and R. Walkling. "A Test of the Free Cash Flow Hypothesis: The Case of Bidder Returns." J. Finan. Econ. 29(1991):315­35.

Lee, H., and R.G. Chambers. "Expenditure Constraints and Profit Maximization in U.S. Agriculture." Amer. J. Agr. Econ. 68(1986):857­65.

Lehn, K., and A. Poulsen. "Free Cash Flow and Stockholder Gains in Going Private Transactions." J. Finance 44(July 1989):771­87.

Miller, L.H., P.N. Ellinger, P.J. Barry, and K. Lajili. "Price and Nonprice Management of Agricultural Credit Risk." Agr. Fin. Rev. 53(1993):28­41.

Nasr, R.E. "Capital Structure and Efficiency in Illinois Grain Farms." Unpub. Ph.D. dissertation, Dept. of Agr. and Consumer Econ., University of Illinois, Urbana-Champaign, March 1996.

Neff, D.L., P. Garcia, and R.H. Hornbaker. "Efficiency Measures Using the Ray-Homothetic Function: A Multiperiod Analysis." S. J. Agr. Econ. 23(December 1991):113­21.

Novak, M.P., and E.L. LaDue. "Stabilizing and Extending Qualitative and Quantitative Indicators of Credit-worthiness in Agricultural Credit Scoring Models." Agr. Fin. Rev. 57(1997):39­52.

Seiford, L.M. "Data Envelopment Analysis: The Evolution of the State of the Art, 1978­1995." J. Productivity Anal. 7(July 1996):99­138.

Splett, N.S., P.J. Barry, B.L. Dixon, and P.N. Ellinger. "A Joint Experience and Statistical Approach to Credit Scoring." Agr. Fin. Rev. 54(1994):39­54.

Tauer, L. "Age and Farmer Productivity." Rev. Agr. Econ. 17(1995):63­69.

Tauer, L., and K.P. Belbase. "Technical Efficiency of New York Dairy Cows." Northeast. J. Agr. and Resour. Econ. 15(1985):10­16.

Whittaker, G.W., and M.J. Morehart. "Measuring the Effect of Farm Financial Structure on Efficiency." Agr. Fin. Rev. 51(1991):95­105.

 

<top>

 

---

Send questions and comments to fsb1@cornell.edu

This page was last modified on: 04/07/04