Testing Sensitivity of Major Market Stock to Oil and Gas Prices in the US

Introduction

In the United States of America (USA), firms having stock prices that are so much sensitive to the changes in the international prices of oil and gas tend to adopt open positions in their goods derivatives as well as to expose the unique values at risk (VAR). Such stock also tends to express their derived contract sensitivity to changes in other market prices. However firms having stock prices that are not so sensitive to the international prices of both oil and gas tend not to adopt open positions in their goods derivatives unless through simplified tabulations of the specific derivative contract (Mian, 1996). Therefore, the firms that are more prone to risk are the ones that are highly expected to use stock sensitivity derivatives to mitigate their risks. It is also advisable for such firms also to voluntarily disclose their updated stock reports in derivative contracts as required by the Securities & Exchange Commission (SEC)’s report (Rajgopal, et al 1998).

Stock sensitivity tests reveal that the required information has been disclosed to the market, with the firms that complied to disclose variance estimates or sensitivity recording greater advantage over the others in their stock prices sensitivity with that of both oil and gas. It is notable that those firms that declared their derivations as having zero overall stock sensitivity to the general market prices recorded the highest shift (McAnally, 1996). This demonstrates the significance of the firm’s disclosure of their derivative contract, in shaping the investor’s attitudes towards the firm’s position in respect to market risks.

This paper examines the linkages between the United States (U.S.) firm’s sensitivity of stock prices to their prevalent risk factors and the industrial risks disclosed by the 1997 SEC’s report.

The SEC Report on Stock Sensitivity Tests

The SEC report underlined the significance and set as mandatory requirement for the US firm’s to include the perceived market risk and stock sensitivity reports along with their annual financial reports. The inclusions were found to be in the public interest in that they protect investors, enhance operational efficiency, encourage creativity and promote capital development. The implementation of the SEC rules was done after a thorough cost-benefit analysis on the mandatory impacts of the stock sensitivity tests on the mainstream firms (Linsmeier, et al 1997). The firms would be at a better position once the investors discover that their derivatives are not speculative but are actual representations of market risks and an important tool to develop more effective strategies for internal risk mitigation (Tufano, 1998).

SEC offered to reduce on the $40 million logistical budget for compliance to stock sensitivity tests in the US through the following two ways that significantly affects the execution of this paper;

Size exemption

The small and medium firms were given a full year to be fully compliant with the stock sensitivity tests. This paper is focused on the initial year when most of the firms presented their reports. The sample of this study is based on the firms left out by the size exemption provisions as the smaller firms took time to comply.

Flexibility provisions

The Securities & Exchange Commission provided that the companies could use their own unique ways in disclosing their findings from the stock sensitivity tests (Kane, et al 1977). In this way it was possible to reduce the logistical costs in that every firm would tailor their disclosures in line with their usual risk management strategies. The flexibility was also a way of making the content more relevant in that the information regarding the tests was more unique and based to the firm’s operations (French, et al 1983). Though the disclosures were made flexible the firms were supposed to adhere to three main guidelines;

• Presentation of fair and well tabulated information on the stock sensitivity tests which were relevant to the eventual determination of the company’s cash flows as dictated by the prospected maturity dead lines.
• Inclusion of an in-depth stock sensitivity analysis showing clearly the possible indications in future losses from the company’s total earnings and a fair representation of the values from an exemplified foreseeable variation in general market prices.
• VAR representations showing any possible foreseeable future losses in the company’s earnings based on the dynamic nature of the market within a specified time frame, a stated level of likelihood and the perceived effect of the consequence to the company.

Methodology

This paper is based on case study analyses of sample US mainstream companies. Stock sensitivity to oil and gas prices as contained by the disclosed tabulated reports from these mainstream companies is studied. Based on the available data before the implementation of the SEC report in USA, it is clear that the investors’ perception is significantly affected by the stock sensitivity to variations in oil and gas prices (Fama, et al 1977). It has been found that the actual stock sensitivity and the disclosed tabulations are different. This emphasizes the need for detailed stock sensitivity tests in a firm as exemplified in this paper for such calculations discloses more inherent information that other methods may fail to reveal. The evidence of the relationship between the company’s revealed reports on their respective stock sensitivity tests and the perceived underlying market risks is well outlined. In addition the relationship between the historical estimation methods of sensitivity to oil and gas prices and the modern ways of actual calculation of sensitivity tests is drawn to show how the latter examines the possible company risks (Venkatachalam, 1996).

Stock sensitivity tests; Data tabulation

Table 1

The following table shows the number of companies considered in the current study and the elimination criteria that was based on the SEC’s provisions. The firms that fulfilled the required criteria are then used in the computation of the stock sensitivity tests. The data used considers the status of the companies as of 28^th January 1997.

From the table it is observable that only 501 companies were successful to meet the requirement on the market capitalization, of which 10 could not be traced in the available Edgar database and therefore their stock sensitivity reports could not be accessed. Fifteen other companies suffered takeovers well before they could be filed whereas fifty eight financial companies and twenty eight insurance based institutions were removed from the list because they had been carrying out their stock sensitivity tests before SEC released its directives. A further twelve companies were removed from the list after undergoing significant structural adjustment programs during the period of study that are believed to have caused serious shift in the level of their risk situation (DeMarzo, et al 1991). Three hundreds and seventy eight companies are therefore left for consideration in the study of which thirty nine are energy related companies that records significant sensitivity in its stock to changes in both oil and gas prices.

Table 2

Table two above shows the data of the energy companies in which thirty nine showed commitment to the use of market goods derivatives but only thirty three of them were found to be active in using these finding in the annual computation of their stock sensitivity tests (Fama, 1965). Twenty two of the thirty nine firms reported their use of sensitivity or value at risk when compared to the overall market commodities risks. Twenty two companies specified oil price as their main exposing factor to market risk while twenty five others mentioned the factor as gas price related. It can be concluded from the table that the other companies could also be exposed to the same market risks.

Table 3 below reveals that the filing period of the above companies was concentrated on a single month, that is, March 1998; however the stock sensitivity tests showed later that it was impractical that such clustered one month period could have any significant effect on the result of the study.

Table 3

Research Design

This paper is meant to test for the linkages between companies’ derivative representations and the stock price sensitivity to market items price; it also evaluates the variations in sensitivity after the filing period of the sample data. To compute these tests an estimation of the perceived shifts in the gradient co-efficient linking oil and gas earnings to the company’s overall equity returns (Wong, 1998). The said tests controls the degree in which the perceived shifts in each company’s stock sensitivity relate to the overall market variations.

Regression estimations for both oil beta and gas beta, that is, the company’s market evaluation of the stock price sensitivity to the contemporary market prices (DeMarzo, et al 1995). The sensitivity is an estimate of the gradient co-efficient of company by company regression percentage shift in the contemporary market price to percentage shift in the company’s stock price over a specified period of time. The equation below is a representation of the returns evaluated as a function of market performance, as based on the percentage variations in the price of oil and the similar variations in the price of natural gas prices.

r_{it =} α_i + β_1ir_mt + β_2ir_ot + β_3ir_gt + ε_it Equation 1

Where, t is the period, r_it is the rate of return on the company’s stock, i is the stock, r_mt is the indication on the returns of the specified market stock, r_ot is the % variation in the oil costs while the r_gt is the similar % variation in gas costs.

Beta co-efficient computes each company’s stock sensitivity based on the performance of the overall market shifts (β_1i), oil costs shifts (β_2i) or natural gas costs variations (β_3i). Therefore an empirical form of equation 1 based on the stock turn over rate based on the data availed within a given time t, can be given by;

r_{it =} α_i + (β_1i/It)r_mt + β_2i /I_t) r_ot + (β_3i /I_t) r_gt + ε_{it ­} Equation 2

Where I_t is the representation of the data accessible by investors within the duration t.

It can be noted that the co-efficient β_i, in the equation 2 above shows that the market based sensitivity for each of the company’s stock fall back to percentage shifts in the contemporary market value of the oil cost and natural gas cost. This co-efficient naturally depends on the data used by the market dynamics in assessing the sensitivity. Secondly it is notable that H1: that is, β_2i, the perceived oil cost sensitivity for company-i, and β_3i the perceived natural gas cost sensitivity for company-i, are inter linked as to whether the companies are making quantitative representation to generalized market risk.

Considering that the majority of the companies make quantitative representation of their market derivatives rather than their position representation, support for H1 would imply that companies that are more exposed to oil and gas costs risk are making more use of derivatives to control market risk factors (Bhattacharya, et al 1999). H1 is also supported consistently by the companies that tend to comply with SEC directives because such companies are more likely to be keen to market price risk and they relay such comprehensive company’s report to the investors in their periodic report statements.

H2: the post representation change in β_2i, that is, the assessed market state of oil cost sensitivity of company-i and β_3i, that is, the assessed market state of gas cost sensitivity of company-i are inter linked as to whether the companies are making quantitative representation to generalized market risk (Collins, et al 1996).

H2 receives support if first, quantitative representation sends new data that may imply that the company is using and all round effective means to evaluate and manage perceived market risks. This can influence the investor’s general perception on the company’s sensitivity to market risk for a long period for as long as the said market risk methodologies and policies are in use. Secondly, the representation sends an alert on the company’s sensitivity to specified market annual position which can only be understood by the management. This alert can however have meaning only to specified annual contracts, unless if the investors would have the managers extend the company’s profile after the expiry of the contractual period (Belsey, et al 1980).

Stock Sensitivity Estimation Procedures

Below is the empirical form of the equation 2 above with the inclusion of other variables that carry out direct testing of the changes in the perceived market risks after the sample’s filing dead line.

r_{it =} α_it + β_1i r_mt + β_2i r_ot + β_3i r_gt + α_2i P_it + β_4i (P_it^*r_mt) + β_5i (P_it ^*r_ot) + β_6i (P_it^*r_gt) + ε_{it ­} Equation 3

The r-variables are the same as before. P_it takes the value of 1 if the period in company-i is within the filing period. In case the period has expired the variable P_it takes the value of zero. The table 4 below shows the tabulated estimate measures of the stock sensitivity indexes computed using the equation 3 above. It should be noted that the overall market beta as well as the oil and gas betas are all taken to be positive for all the companies’ samples. However sample wide changes in both oil and natural gas stock sensitivity are negative.

Table 4

Stock Sensitivity Analysis

Whenever there is lack of exhaustive and properly developed theories that define the reasons for disclosing quantitative data by companies and the relevant significance that such data would have on stock sensitivity, it is always advisable to carry out corresponding tests so as to authenticate the findings (Beaver, et al 1970). For example in our finding, a regression to the mean can be a proof the company with the greatest square of either the oil or gas beta experiences more change in the squared betas as a result to representation of the given input data. Therefore, two different modes of stock sensitivity tests are being considered here, these are, sensitivity based on the outcome to the sample selected and the sensitivity based on the outcome to the specification of the sample time line. In the oil based samples, the variables are like those seen in the complete sample; however, reduction in the importance of the statistic results means a responsive reduction in the size of the sample. Similarly in the natural gas sample, the absolute size and the importance of the coefficient estimations in the level of the gas together with the changes in the gas model may reduce but the direction is always maintained. It can therefore be concluded here that the oil intensive part of the main sample is the basic driving force of the entire sample.

A repeat of the analysis by use of pseudo activity dates that were randomly selected within a period of sixty days before and after each company’s actual filing date, revealed a similar change in both oil and gas betas respectively (Beaver, et al 1979). The results could be attributed to the typical regression towards the central mean as well as to the unspecified derivative representation that drive and dictates the shifts in beta estimations. If for example, the same kind of varying timing is used with a company A filing its respective report twenty days before another company B, then its pseudo activities are expected to be twenty days before those of the other company B. When these tests are repeated with the entire sample for both oil and gas part of the samples, the result persists that companies with more beta in oil and gas always would tend to make quantitative representation on the real activity time line. However, changes in the beta of either oil or gas showed no significant response to the pseudo activity time line in all the tests and samples that were used. These lend to a general inference that the changes in beta on the real activity time line from data is contained only in the market risk representation and not to others externally related factors.

Conclusion

The study reveals that companies’ having their stock prices that are highly sensitive to the changes in both oil and gas costs, tends to adopt open positions towards the end of the year in regard to market derivatives and release their VAR or sensitivity towards their respective contractual derivative on contemporary market prices. For the companies with stock prices that are not highly sensitive to both oil and gas costs, tend not to naturally make use of market commodity derivatives and in case they do, they offer little information on contractual derivatives.

These facts can explain the observations that those companies that are more exposed to price risks are the ones that are more likely to reveal their respective stock sensitivity tests reports (Bartov, 1994). These reports help in shaping the perceptions on their efficiency in the methodologies used in management of risks or the sensitivity of end of year report and degree of accuracy that such stock sensitivity tests measures. In the report such evidence is seen indirectly as the firms sought to comply with the directives of SEC and disclose relevant indicators on their derivative contracts.

Companies that revealed vale at risk (VAR) or sensitivity had a greater advantage than those companies that did not in relation to stock price sensitivity towards both oil and gas costs (Barth, et al 1996). In addition it has been seen that disclosures of stock sensitivity tests by the companies along with their annual company reports helps in building the confidence of the investors on the ability of the company to manage foreseeable risks and mitigate its effect. Stock sensitivity tests is therefore a significant tool that firms, companies and industries must adopt, regardless of their sizes and use it along with other market risk assessment tools to improve on their preparedness to market commodity risks and in building positive attitude in investors over the running of those firms.

Reference

Barth, M. E.,. Beaver, W. H, and Landsman, W. R.. (1996). Value-relevance of banks’ fair value disclosures under SFAS No. 107. The Accounting Review 71: 513–537.

Bartov, E.and Bodnar, G. (1994). Firm valuation, earnings expectations, and the exchange-rate exposure effect. The Journal of Finance: 1755-1785

Beaver, W. and J. Demski (1979) “The Nature of Income Measurement,” Accounting Review LIV, pp. 38-46.

Beaver, W.,. Kettler, P, and. Scholes, M. (1970). The association between market-determined and accounting-determined risk measures. The Accounting Review: 654-82.

Belsey, D., Kuh E., and Welsch R.. (1980). Regression Diagnostics, John Wiley and Sons, New York, New York.

Bhattacharya, U., H. Daouk, B. Jorgenson, C. and Kehr H..(1999). When an Event is Not an Event: The Curious Case of an Emerging Market, Journal of Financial Economics, forthcoming.

Collins, D. W., and Venkatachalam M.. (1996). Derivatives disclosures and interest sensitivity of commercial banks. Working paper, Stanford University.

DeMarzo, P., and. Duffie D. (1995). Corporate incentives for hedging and hedge accounting. Review of Financial Studies 8 (3): 743–772.

DeMarzo, P.M. and Duffie D., (1991). “Corporate financial hedging with proprietary information,” Journal of Economic Theory, 53, 261-286.

Fama, E. (1965). The Behavior of Stock Market Prices. Journal of Business, 38, No. 1, 34 – 104.

Fama, E. and Schert, G (1977). “Asset Return and Inflation. Journal of Financial Economics, 5, No. 2, 115 – 146.

Financial Accounting Standards Board (1991a). Recognition and Measurement of Financial Instruments, Discussion Memorandum. Norwalk, CT: Financial Accounting Standards Board.

Financial Accounting Standards Board (1991b) Hedge Accounting: An Exploratory Study of the Underlying Issues, Research Report. Norwalk, CT: Financial Accounting Standards Board.

French, K, Ruback, R and Schwert, G. (1983) Effects of Nominal Contracting on Stock Returns. Journal of Political Econimy, 91, No. 1, 70 – 96.

Kane, E. J. and Undal, H. (1977) Change in Market Assessments of Deposit Institution Riskness. Journal of Financial and Quantative Analysis, 12, No. 3, 363 – 346.

Kessel, R.A. (1956) Inflation Caused Wealth Redistribution: A Testof a Hypothesis. American Econimic Review, 46, No. 1, 128 – 141.

Linsmeier, T. J., and Pearson N. D.(1997). Commentary: Quantitative disclosures of market risk in the SEC release. Accounting Horizons 11: 107–135.

McAnally, M. L. (1996). Banks, risk, and FAS No. 105 disclosures. Journal of Accounting, Auditing, and Finance 11: 453–496.

Mian, S.L., (1996). “Evidence on corporate hedging policy,” Journal of Financial and Quantitative Analysis,” 31, 419-439.

Morrison, D.E. and Henkel, R.E. (1970). The Significance Test Controversy. Chicago, Ill.: Aldine Publishing Company.

Nance, D.R.,. Smith C.W, and. Smithson C.W, (1993) “On the determinants of corporate hedging,” Journal of Finance,” 48, 267-84.

Rajgopal, S. (1998). The Association between Accounting Disclosures and the Net Commodity Price Risk Exposure of Oil and Gas Producers. Working paper, University of Washington.

Rajgopal, S. and M. Venkatachalem (1998). The Association Between Earnings Sensitivity and Market Determined Risk Exposures: The Case of Oil Price Risk for Petroleum Refiners. Working paper, University of Washington and Stanford University.

Ritchken, P. (1996). Derivative Markets. New York: Harper Collins.

Smith, C.W. and Stulz, R.M. (1985). “The determinants of firms’ hedging policies,” Journal of Financial and Quantitative Analysis, 20, 391-405.

Titman, S and Warga, A. (1989). Stock Returns as Preditors of Interest Rates and Inflation. Jornal of Financial and Quantitative Analysis, 24, No 1.

Tufano, P. (1998). The determinants of stock price exposure: Financial engineering and the gold mining industry. Journal of Finance: 1015-1052 33

Venkatachalam, M. (1996). Value Relevance of banks’ derivatives disclosures, Journal of Accounting and Economics, 327-355

William P. L. and Richard A. S. (1977). A test of stone’s Two index Model of Returns. Journal of Financial and Quantitative analysis, 12, No. 3, 363 – 346.

Wong, M. H. F. (1998). The usefulness of SFAS 119 derivatives disclosures for assessing the foreign exchange risk exposure of manufacturing firms. Working paper, University of California at Berkeley.