(64d) Integrating Finance and Control for Process Operations

This paper considers prospects for an integrated approach to the finance and control of processes producing commodity goods. Financial markets offer producers opportunities to reduce short term exposure to many market uncertainties. To Hedgers take positions in financial instruments, such as futures and options, that offset the impact of market volatility. In particular industries, such as electricity generation and commodity agriculture, the practice is so well refined that producers value and operate their production assets as if they were financial instruments. The return to the producer is the net result of integrated financial and process operations. The goal of this paper is to describe general principles underlying these practices, and to consider and compare analytical approaches.

Why hedge process operations? In spite of the wide spread use of hedging1, riskiness, by itself, is not a sufficient reason to hedge process operations. In a frictionless capital market, the Capital Asset Pricing Model (CAPM) and the ModiglianiMiller theory suggest that large, publicly held firms gain no value from hedging designed solely to reduce riskiness. This is because shareholders can eliminate diversifiable risk by managing a portfolio of investments, thus there is no additional value to the investors for the firm to hedge diversifiable risk. Similarly, shareholders can allocate wealth between riskless assets and the market (the ?two-fund' theorem), so again they receive no additional value from a public firm hedging solely to reduce market risk.

Recent empirical work [13, 4], supported by theoretical models [23, 7, 2, 11, 22, 9, 19, 5] shows firms, due to market inefficiencies and frictions hedge, hedge in order to: (1) avoid financial distress, (2) minimize the impact of tax convexities, (3) reduce the need for outside capital, and (4) maintain a more stable cash flow with which to pursue other value generating opportunities. Further, investors whose wealth cannot be diversified are typically riskadverse. Taken together, these reasons suggest that value can be returned to investors from operations by reducing cash flow volatility.

Consistent with CAPM and other capital pricing models, the objective of the integrated control of finance and process operations is to operate at the ?efficient frontier,' that is to minimize cash flow volatility for a given expected rate of return. Conceptually, we seek methods to dynamically control a process and an associated hedge portfolio in response to market and process observations. The goal is to maximize an objective function that includes a weighted measure of expected return and cash flow volatility.

Relevant prior work includes Anderson and Danthine's [1, 21] consideration over a single period of the production of multiple cash goods subject to price uncertainty. By introducing a hedge against correlated futures contracts, they found a optimal hedge position and production levels. Empirical application to a cattle feeding operation [20] demonstrated enhanced return and reduced variance compared to traditional hedging strategies and the existence of an ?efficient frontier' of operation. In a major paper, Brennan and Schwartz [3], constructed a dynamic hedge for a hypothetical mining operation and coupled control strategy that provided for mine startup, shutdown, and abandonment based on observed price signals. A far reaching result of this approach was to establish a 'real option' value for the optimally operated process. Other relevant work includes [22, 14, 10, 6, 15]. In particular, a recent theoretical problem demonstrates application of these ideas to inventory management [12].

This work is closely related to the rapidly growing literature on socalled 'Real Options'. In particular, establishing risk minimizing hedge portfolios and stochastic dynamic programming are key methods. The Real Options literature, unfortunately, is remarkably lax in discriminating among sources of diversifiable and idiosyncratic risks, describing decision processes, and in the application of the socalled Marketable Asset Disclaimer (MAD) [8]. For the applications envisaged here, the resulting cashflow streams are not likely to be in the span of cash streams produced by marketable commodity derivatives. Thus Luenberger's work on pricing [17, 18] for incomplete markets is an essential element of this work.

Also of direct relevance to this work are computational techniques for larger scale and approximate stochastic dynamic programming [16] including the more recent work by Tsitsiklis and Van Roy, and Jay Lee.

To fix ideas, the paper will consider the valuation of a novel ?Ethanol Crack Spread.' Preliminary work has demonstrated an alternative commodity spread for Ethanol based on weighted sum of CBOT Ethanol, Corn, Gas, and Soybean futures. The spread is a mean-reverting subject to significant volatility. The consequences for ethanol process operation, management, and valuation will be discussed.

------------

1) In a 1990 survey of the 372 non-financial firms among the Fortune 500, Geczy. et al. [13], found 59.1% of all firms used financial derivatives of some form. That fraction increased to 68.8% among energy producers, and to 75% in petroleum refining.

References

[1] Ronald W. Anderson and JeanPierre Danthine. Hedging and joint production: Theory and illustration. The Journal of Finance, 35(2):487?498, May 1980.

[2] Hendrik Bessembinder. Forward contracts and firm value: Investment incentive and contracting effects. The Journal of Financial and Quantitative Analysis, 26(4):519?532, 1991.

[3] Michael J. Brennan and Eduardo S. Schwartz. Evaluating natural resource investments. The Journal of Business, 58(2):135?157, 1985.

[4] Gregory W. Brown, Peter R. Crabb, and David Haushalter. Are firms successful at selectively hedging? preprint, February 2003. online at http://papers.ssrn.com/sol3/papers.cfm? abstract id=385169.

[5] Gregory W. Brown and Klaus Bjerre Toft. How firms should hedge. Review of Financial Studies, 15(4):1283?1324, 2002. available at http://ideas.repec.org/a/oup/rfinst/ v15y2002i4p12831324.html.

[6] Renxe Caldentey and Martin Haugh. Optimal control and hedging of operations in the presence of financial markets. preprint, 2004. available online at http://bctim.wustl.edu/calendar/ mediafiles/ACF1BCF.pdf.

[7] Tim S. Campbell and William A. Kracaw. Corporate risk management and the incentive effects of debt. The Journal of Finance, 45(5):1673?1686, December 1990.

[8] Tom Copeland and Vladimir Antikarov. Real Options: A Practitioner's Guide. Thomson Texere, 2003.

[9] Peter M. DeMarzo and Darrell Duffie. Corporate incentives for hedging and hedge accounting. The Review of Financial Studies, 83(3):743?771, Autumn 1995.

[10] Michael Denton, Adrian Palmer, Ralph Masiello, and Petter Skantze. Managing market risk in energy. IEEE Transactions on Power Systems, 18(2):494?502, May 2003.

[11] Kenneth A. Froot, David S. Scharfstein, and Jeremy. C.steinrisk management: Coordinating corporate investment financing policies. The Journal of Finance, 48(5):1629?1658, December 1993.

[12] Vishal Gaur and Sridhar Seshadri. Hedging inventory risk through market instruments. preprint, available online at http://pages.stern.nyu.edu/vgaur/HedgingUnabridged% 2020041027.pdf, 2004.

[13] Christopher Geczy, Bernadette A. Minton, and Catherine Schrand. Why firms use currency derivatives. The Journal of Finance, 52(4):1323?1354, September 1997.

[14] Bardia Kamrad and Ricardo Ernst. An economic model for evaluating mining and manufacturing ventures with output yield uncertainty. Operations Research, 49(5):690?699, 2001.

[15] Bardia Kamrad and Keith Ord. Market risk and process uncertainty in process operation. In 9TH Annual International Conference on Real Options, 2005. online at http://www. realoptions.org/submission/papers2005/REAL OPTINS MPR IN PRODOPS JAN 2005.pdf.

[16] Francis A. Longstaff and Eduardo S. Schwartz. Valuing american options by simulation: A simple leastsquares approach. The Review of Financial Studies, 14(1):113?147, 2001. Papers and Proceedings FiftySeventh Annual Meeting, American Finance Association, New Orleans, Louisiana January 46, 1997.

[17] David G. Luenberger. Projection pricing. Journal of Optimization Theory and Applications, 109(1):1?25, 2001.

[18] David G. Luenberger. A correlation pricing formula. Journal of Economic Dynamics & Control, 26:1113?1126, 2002.

[19] Antonio S. Mello and John E. Parsons. Hedging and liquidity. The Review of Financial Studies, 13(1):127?153, Spring 2000.

[20] Mikhail A. Noussinov and Raymond M. Leuthold. Optimal hedging strategies for the u.s. cattle feeder. Technical Report 9804004, Economics Working Paper Archive at WUSTL, Apr 1998. available at http://ideas.repec.org/p/wpa/wuwpfi/9804004.html.

[21] Gordon C. Rausser. Hedging and joint production: Theory and illustrations: Discussion. The Journal of Finance, 35(2):498?501, May 1980.

[22] Yong Sakong, Dermot J. Hayes, and Arne Hallam. Hedging production risk with options. American Journal of Agricultural Economics, 75:408?415, May 1993.

[23] Clifford W. Smith and Rene M. Stulz. The determinants of firms' hedging policies. The Journal of Financial and Quantitative Analysis, 20(4):391?405, December 1985.