Crude Oil Price as a Determinant of Industry Stock Returns

CHAPTER 1: INTRODUCTION

Crude Oil is the crucial input of modern economies. As countries urbanize and renovate their demand for oil raises drastically. Potential demand for oil is hard to forecast but is usually highly correlated with the growth in industrial production. Therefore, countries experiencing hasty economic growth are the ones probably to significantly amplify their demand for crude oil. Increases in oil demand without equalizing increases in supply lead to higher crude oil prices. Higher crude oil prices act like an inflation tax on consumers and producers by:

1) plummeting the amount of disposable income consumers have left to spend on other goods and services, and

2) increasing the costs of non-oil producing companies and, in the absence of fully passing these costs on to consumers, sinking profits and dividends which are key drivers of stock prices.

In addition to worldwide demand and supply conditions, crude oil prices also respond to geopolitics, institutional arrangements (OPEC), and the dynamics of the futures market (Sadorsky, 2004). Unanticipated changes in any of these four factors can create volatility, and hence risk, in oil futures prices. Oil Price volatility increases risk and uncertainty which negatively impacts stock prices and reduces wealth and investment.

One macroeconomic factor that is receiving increasing empirical attention is crude oil. A key factor input, crude oil prices have the potential to dramatically alter the financial performance of national economies and the firms that operate therein. it is reasonable to expect that stock markets are profoundly influenced by oil price changes, remarkably little empirical evidence exists. Sadorsky (2001) argues that there has been a large volume of work investigating the links among international financial markets, and some work has also been devoted to the interaction among crude oil spot and future prices.

In contrast, little work has been done on the relationship between oil spot/futures prices and stock indices. Even the findings of the extant work are mixed. Poon and Taylor (1991) found no evidence of an oil price factor in the U.S. and Japan, respectively. In contrast, Sardorsky (1999) concluded that oil prices were a significant factor in the U.S. Jones and Kaul (1996), Faff and Brailsford (1999), Sardorsky and Henriques (2001), and Sardorksy (2001) have also examined the impact of oil price factors with disparate results. While these studies have provided at least some evidence that oil prices constitute a source of systematic asset price risk, and that the exposure to this risk varies across industries, no recent work is known in the Pakistani context.

Statement of Problem

At least since the development of the capital asset pricing model, a literature has sought to identify the determinants of asset prices and returns. Given the capital asset pricing model rests on the premise that assets are priced according to their covariance with the market portfolio, the increasing acceptance that other pricing factors, especially macroeconomic factors, should also be modeled has led to yet further refinements, most notably in the form of the arbitrage pricing theory. With this multifactor specification as a starting point, an increasing number of empirical studies have sought to investigate whether macroeconomic variables constitute a source of systematic asset price risk at the market and industry level ((Antoniou et al. (1998), Faff and Chan (1998), Canova and Nicolo (2000)). The fundamental endeavor of this analysis is to find out whether macroeconomic information, particularly crude oil prices, gives incremental information beyond the market portfolio about the behavior of industry stock returns

Hypotheses

Crude oil being the core input of productions it has been assumed that changes in crude oil price significantly changes the cost of production. Therefore an increase in crude oil price leads to higher cost of production. Consequently higher cost of production leads to lower profit margins or it forces producer to increase the price of the goods. And increase in price of goods leads lower demand for the good resultantly sales of the firm goes down and overall profitability suffer. Further more firm making tiny or negative profit loses investors’ confidence and its stocks price go down which leads to negative stock returns and vice versa. Following hypotheses are suggested:

H1: change in oil prices has significantly impact on the stock returns of Automobile and Parts sector of Pakistan

H2: change in oil prices has significantly impact on the stock returns Energy sector of Pakistan

H3: change in oil prices has significantly impact on the stock returns Chemical and Pharmaceutical sector of Pakistan

H4: Change in oil price has different impact on the stocks return of different industrial sectors.

CHAPTER 2: LITERATURE REVIEW

Asset prices are generally believed to respond sensitively to macroeconomic news. Every day experience gives the impression to support the observation that individual asset prices are influenced by a broad range of unpredicted events and that various events have a more persistent impact on asset prices than do others (Faff and Chan, 1998). Therefore macroeconomic news is important factor in the explanation of stock returns at the industry level.

In recent years there have been numerous studies which argued that stock prices not only replicate changes in current and future cash flows and anticipated returns, but are also determined by speculative dynamics that is investor attitude and/or overreaction to news. Many researchers have claimed that the strong predictability of stock returns over various horizons is proof of such fads. In an endeavor to measure whether the predictability of stock returns is rational, several recent studies tested whether using Capital Asset Pricing Model (CAPM) or a more general asset pricing model like the Arbitrage Pricing Theory (APT) could eliminate or ex-plain their predictability. If factors and/or their coupled risks can explain the predictability of stock returns then the market is convincing, and vice versa (Fama and French, 1989).

The approach taken in this paper uses a global multi-factor model that permits for both unconditional and conditional risk factors. This approach is related to the international capital asset pricing model (CAPM), the implications of which have been studied by Brealey & Myers, (2002). Whereas the focus of the CAPM is on market risk, the multi-factor model includes multiple sources of risk (Ross, 1976). The CAPM and multi-factor models are essential building blocks of contemporary portfolio theory. In both models, expected returns are linearly connected to risk factors and risk premiums. So far the CAPM has been broadly tested both domestically and internationally and the general agreement is that the CAPM explains no statistically significant correlation between systematic risk (beta) and returns (Berk, 1995).

Modern economies are more energy efficient nowadays than they were 40 years ago with oil usage per dollar of GDP less than half of what it was in the 1970s. This increase in energy efficiency has happened because of cheap energy intensity through technological modernization and more dependence on a broadened range of energy sources (like a greater mix between non-renewable and renewable energy sources). Emerging and new economies tend to be more energy intensive than more developed economies and are therefore more exposed to high oil prices. Consequently, oil price changes are likely to have a larger impact on earnings and stock prices in emerging economies.

Past practice has shown that oil price shocks have a much bigger impact on the poorer countries in the world. The OPEC oil embargo of 1973, which raised the price of oil from $3 per barrel to $13 barrel in just over a few short months, created real economic and social destitution for developing countries by increasing their costs of imported crude oil. Worldwide lending institutes like the World Bank and the International Monetary fund (IMF) had to grant loans to developing countries so that they could keep on with their economic development projects (Canova, 2000).

If crude oil plays a vital role in an economy, one would anticipate changes in oil prices to be interrelated with changes in stock prices. Specifically, it can be argued that if oil influences real economic activity, it will impact earnings of those businesses in which crude oil is (directly or indirectly) a factor of production. Thus, a swell in crude oil price would causes projected earnings to change, and this would lead to an immediate change in stock prices if the stock market efficiently capitalizes the cash flow propositions of the oil price increases. (Canova and Nicolo, 2000).

Sadorsky (2003) exercised monthly data from July 1986 to April 1999 to examine the macroeconomic determinants of U.S. technology stock price conditional volatility. Sadorsky (1999) projected a vector auto regression model with monthly data to cram the affiliation between oil prices changes and stock returns in the United States. In his analysis, he found that oil price alteration and oil price volatility both play vital roles in affecting stock returns. The pragmatic results indicated that the conditional volatilities of industrial production, oil prices, the default premium, the federal funds rate, the foreign exchange rate, and the consumer price index each have a significant impact on the conditional volatility of technology stock prices.

According to McSweeney and Worthington (2007) surplus returns in the retailing industry are negatively connected to the oil price factor. A latent explanation for the observed negative effect is the influence of oil price increases on consumer discretionary spending. Since the price of oil get higher relative to other goods and as a percentage of household expenditure, the nondiscretionary character of household petroleum expenses, at least in the short-run, restricts the amount of discretionary funds presented to consumers. This ought to lower the returns on retail firms.

Basher and Sadorsky (2006) studied the influence of oil price on 19 emerging equity markets including Pakistan. They found strong proof that oil price risk influences stock price returns in emerging equity markets while the precise relationship depends, to some extent, on the data frequency being used. The conditional association is not yet symmetrical. For daily and monthly data, positive oil price changes have a positive effect on surplus equity market returns in emerging economies. For weekly and monthly data, negative oil price changes have positive and significant effects on emerging equity market returns.

Faff and Brailsford (1999) in their study found that the degree of pervasiveness of an oil price factor, beyond the influence of the market, is detected across some Australian industries, positive oil price sensitivity in the Oil and Gas and Diversified Resources industries and similarly they found significant negative oil price sensitivity in the Paper and Packaging, and Transport industries. Generally, they revealed that long-term effects persist, although they hypothesize that some firms have been able to pass on oil price changes to customers or hedge the risk.

Mohan Nandha, and Faffa (1999) analyzed thirty five global industry indices for the period of 22 years from April 1983 to September 2005.They indicated that oil price climb has a negative impact on stock returns for all industries apart from mining, and oil and gas sectors. Moreover in United Kingdom Idris El-Sharif, Dick Brown, Bruce Burton, Bill Nixon and Alex Russell analyzed that the oil and gas sector, using data concerning to the United Kingdomn (UK), the major oil producer in the European Union (EU). Their findings pointed out that the relationship, all the time, is positive, often highly significant and reveals the direct impact of volatility in the price of crude oil on equity values within the industry.

Additionally, Mcsweeney and Worthington (2007) examined the impact of macroeconomic risk factors on Australian industry returns. Their research indicated that the macroeconomic factor specifically oil prices are important determinants of excess returns for many industries. Of the nine industries considered, the energy industry exhibited a strong positive association with oil price increases, while the banking, retailing and

CHAPTER 3: RESEARCH METHODS

Methods of Data Collection

Secondary data has been collected to scrutinize the relationship between macroeconomic variables and industry stock returns, monthly data over the period July 2003 to June 2008 has been employed. The choice of a monthly frequency is consistent with previous work which examines macroeconomic variables in relation to equity returns (Faff and Brailsford (1999), Sadorsky (2001).

Sample Size

Monthly data of industrial returns, crude oil prices, KSE 100 and foreign exchange rate has been collected for the period of 5 years since July 2003 to June 2008. Data has been gathered from State Bank of Pakistan (SBP), Karachi Stock Exchange (KSE) and OGRA. Automobile and allied, Oil and Gas, Chemical and Pharmaceuticals sectors have been considered to examine the impact of crude oil price on their stock returns.

Research model

The fundamental endeavor of this analysis is to conclude whether macroeconomic information, specifically crude oil prices, gives incremental information beyond the market portfolio concerning the behavior of industry stock returns. While at least some work has been conducted at the market level (Charles, Jones and Gautam, (1996 ), and Antoniou, Garrett and Priestley, (1998)) reasonably few studies have attempted to scrutinize the relationship between macroeconomic factors and stock returns at the sector level. To representation the relationship between the macroeconomic factors and industry returns, a multifactor model following Mcsweeney and Worthington (2007), Faff and Brailsford (1999), and Sadorsky (2001) is employed.

r it = βi0+ βi1 mkt +βi2 oil +βi3 fx + eit

where r it denotes the return on the stock index of the ith industry at time t, mkt is the return on the market portfolio, oil is the change in oil prices, fx is the change in the exchange rate, βi are parameters to be estimated that are expected to vary by industry, and eit is the error term.

Variable

Industrial Stock Returns

Industrial stock return is the monthly yield (return) on the price of the stock of particular industry for a given time; the prices have of stocks have been taken from Karchi Stock Exchange. The stock return in each industry is calculated as:

rit =ln (indi;t/indi;t-1)

where, rit is the continuously compounded monthly return for industry i at time t, indit and indi, t-1 are the index prices for industry i at time t and t -1, respectively.

Oil Price

Oil price is the percentage of oil price change for a particular given time period (one month). The oil price factor is constructed as:

oilt = ln(wtxt/wtxt-1)

Where, oilt is the log monthly change in the oil price at time t, and wtxt and wtxt-1 is the respective price of oil at time t.

Exchange Rate

The exchange rate is the percentage change in foreign exchange (PKR/USD) rate for a given time period (one month). The exchange rate factor is constructed as:

fxt = {ln (PKRt/USDt)/(PKRt-1/USDt-1)}

Where, fxt is the log monthly change in the PKR/USDt exchange rate at time t, and

PKR/USDt-1 is the respective PKR/USD exchange rate at time t and time t – 1

Market Returns

The market stock return is the yield (return) on KSE 100 index for one month period of time. The market return on the market portfolio is calculated as:

mktt =ln (aoiit/aoiit-1)

Where, mktt is the continuously compounded monthly return for the aggregate market Index at time t, aoiit and aoit-1 are the values for the market index at time t and t -1, respectively.

Statistical Technique

Multiple Linear regression technique was use to measure the impact of crude oil price changes on industrial stock returns

CHAPTER 4: RESULTS AND FINDINGS

Results

In this study crude oil price, foreign exchange rate and KSE index as independent variable were used to predict the variation in industrial stock returns such as chemical and pharmaceutical sector, automobile and parts sector and energy sector of Pakistan consistent with the previous study of McSweeney and Worthingto (2007).

Multiple linear regression analysis was used to analyze the impact of oil price (one month lagged price) on various industrial stock returns. Model summary shows that the regression has performed a great job of modeling industrial stocks returns for various industries. Adjusted R2 explains that nearly 50%, 42% and 49% of the variation in industrial stocks returns for chemical and pharmaceutical, automobile and parts and energy sectors respectively is explained by the model. R is the multiple correlation coefficients; it shows the relationship between the independent variables and dependant variable. The values 2.436, 2.133, and 2.716 (> 2.0) of Durbin Watson autocorrelation test show that there is no autocorrelation which may influence the model.

Model Summary

Model	R	R Square	Adjusted R Square	Std. Error of the Estimate
Chemical & pharmaceutical Sector	.709a	.502	.493	24.709050
Automobile & Parts Sector	.647a	.419	.409	81.284967
Energy Sector	.697a	.486	.468	30.752494

The ANOVA table below shows a significant value of F statistic, representing that using these models are better than supposing the mean. This table elaborates that regression output presents information about the deviation accounted for by the model. While, residual explains that information not explained by the model. The model with high regressed value shows that the major deviation explained in the dependent variable by the predictors (independent variables). The f test is the regression mean square upon the residual mean square. The significant value of f test is less than 0.05 for all three models. This indicates that a linear relationship occur between crude oil price and industrial stock returns.

ANOVA Table

Model

Sum of Squares

df

Mean Square

F

Sig.

Chemical & Pharmaceutical Sector

1

Regression

35090.060

1

35090.060

57.474

.000

Residual

34800.617

57

610.537

Total

69890.677

58

Automobile & Parts Sector

1

Regression

266835.984

1

266835.984

40.385

.000

Residual

370005.764

56

6607.246

Total

636841.749

57

Energy Sector

1

Regression

49255.578

2

24627.789

26.041

.000

Residual

52014.375

55

945.716

Total

101269.954

57

Table of Coefficients

Model

Unstandardized Coefficients

Standardized Coefficients

t

Sig.

Collinearity Statistics

B

Std. Error

Beta

Tolerance

VIF

Model						Tolerance
Chemical & pharmaceutical Sector
Oil price	.034	.004	.709	7.581	.000	1.000
KSE100 Index	.764	.745	.102	1.026	.310	.909
Exchange Rate	2.736	6.152	.046	.445	.658	.861
Automobile & Parts Sector
		(Constant)	-11.352	10.858	-1.045	.300
Oil price	.094	.015	.647	6.355	.000	1.000
KSE100 Index	-.856	2.458	-.038	-.348	.729	.909
Exchange Rate	-7.805	20.296	-.043	-.385	.702	.861
Energy Sector
Oil price	.042	.006	.721	7.210	.000	.934
KSE100 Index	1.934	.903	.214	2.141	.037	.934
Exchange Rate	7.972	7.555	.110	1.055	.296	.861

The tolerance is known as the percentage of variance in a given independent variable (predictor) that cannot be contributed by the other independent variables or predictors. Thus, the high tolerances 1.00, 1.00 and 0.934 of predictor oil price for chemical and pharmaceutical, automobile and energy sectors respectively show that no variance in a specified predictor or independent variable can be explained as a result of the other predictors. If the tolerance level is close to 0, it can be said there is a greater multicollinearity and it will inflate the standard error of regression coefficient. A VIF higher than 2.0 is usually considered problematic, and the VIF in the table is all less than 2.00.

Market models augmented by an oil price, exchange rate and KSE 100 index factor are estimated with ordinary least squares over the period July 2003 to June 2008 for each of the three industries. The estimated coefficients, standard errors and p-values of the parameters detailed in Equation (1) are presented in Table of coefficients. Model summary includes the R2, the adjusted R2 from a single-factor market model, and an F-test of the null hypothesis that all slope coefficients are jointly zero and its p-value.

The estimated models are all highly significant at the five-percent level, as indicated by the F-statistics and associated p values. The values of R2 ranges between 0.647 (Automobile & Parts Sector) and 0.709 (Chemical & Pharmaceutical Sector), indicating that between 65 and 70 percent of the variation in excess industry stock returns is accounted for by the models. Hence, the models appear to fit the data relatively well.

The constant term in two (Energy and Chemical and Pharmaceutical sector) estimated models are insignificant and it is significant for automobile and parts sector. The statistical insignificance of the constant term is consistent with previous empirical studies of stock returns and macroeconomic factors (Faff and Brailsford (1999), Sadorsky and Henriques (2001)).

In terms of the sensitivity of Pakistani industry returns to the oil price factor, the estimated coefficient (in brackets) is significant in all three models; Chemical & Pharmaceutical (0.034), Energy (0.042) and Automobile and Parts (-0.094) sectors.

Excess returns in the automobile industry are positively related to the oil price factor. A possible explanation for the observed positive effect is the ability of the automobile sector to transfer the increase in cost due to increase in the price of inputs (oil price) to consumer by increasing the price of vehicles. This can increase the returns of the sector.

Previous empirical evidence suggests that the association between exchange rates and stock returns is both country and industry specific. The estimated regressions indicated that the coefficients for the AUD/USD exchange rate are insignificant for all three sectors.

Hypotheses Assessment Summary

The hypothesis of the study was to identify the change in oil prices has significant of the industrial stock returns of automobile, chemical and pharmaceutical and energy sectors of Pakistan. This table shows the statistical result about the rejection and acceptance of the hypotheses.

TABLE: Hypotheses Assessment Summary

S.NO.	Hypotheses	SIG.
H1	Change in oil prices has significantly impact on the stock returns of Automobile and Parts sector of Pakistan	.000
H2	Change in oil prices has significantly impact on the stock returns Energy sector of Pakistan	.000
H3	Change in oil prices has significantly impact on the stock returns Chemical and Pharmaceutical sector of Pakistan	.000

The hypothesis that change in oil price has significant impact of the stock returns of automobile and parts, energy, and Chemical and Pharmaceutical sectors have been accepted at 95% confidence and sig. value of .000. It showed that change in crude oil price significantly influence the industrial stock return of all three industries. Further more, the beta coefficients of oil price of three industries are not equal (Automobile and Parts (0.094), Energy (0.042) and Chemical and Pharmaceutical (0.034)) showed that the impact of change in crude oil price effect differently on stock returns of these industries.

CHAPTER 5: CONCLUSION AND DISCUSION

Conclusion

This study examined the impact of macroeconomic risk factors on Pakistani industry stock returns. Multiple linear regressions indicates that macroeconomic factor is specially oil prices, important determinants of excess returns for automobile, chemical and pharmaceutical and energy sectors of Pakistan. Of the three industries considered, all industry exhibited positive significant association with oil price increases. While the negative oil price coefficient were expected for the automobile industry but the significantly positive coefficient for the automobile and part industry is a surprising finding. We suggest that this is because of the ability of transferring the cost burden on the customer of the industry as it seems to have inelastic demand for automobiles in Pakistan. Moreover the oil price coefficient of all three sectors is nearly equal so the hypothesis that oil price has different impact on the industries under study stood not accepted.

Implications and Limitations

This study helped various investors, management and other researcher in analyzing and observing the behavior stocks returns in various sectors against the fluctuation in crude oil prices. Research students should further work on multifactor asset pricing model and study other macroeconomic variables that may influence the stock returns. In this analysis the major issue faced was availability of the data.

Recommendations

This research was limited to the only three sector of Karachi Stock Exchange of Pakistan. The data were taken from July 2003 to June 2008 due to data availability constraint. It suggested that such type of study should be carried out with a large sample size and in other countries of Asia as well, as to have comprehensive idea about oil price fluctuations and stock returns. Moreover, it also suggested that other factors except ones examined in this study should be researched as to have perfect idea about stock returns behavior. Besides that, this study can also be replicated in other developing countries.

REFERENCES

Antoniou, A., Garrett, I., and Priestley, R. (1998). Macroeconomic variables as common pervasive risk factors and the empirical content of the arbitrage pricing theory. Journal of Empirical Finance, 5(3), 221-240.

Asprem, M. (1989). Stock prices, asset portfolios and macroeconomic variables in ten European countries. Journal of Banking and Finance, 13(4-5), 589-612.

Basher, S.A. & Sadorsky, P.(2006). Oil price risk and emerging stock markets. Global Finance Journal.

Berk, Jonathan B. (1995). A Critique of Size-Related Anomalies,” The Review of Financial Studies 8, 275-286.

Brealey and Myers, (2002). Corporate Finance: Capital Investment and Valuation. McGraw-Hill; 1, 109-131.

Canova, F., and Nicolo, G. D. (2000). Stock returns, term structure, inflation, and real activity: An international perspective. Macroeconomic Dynamics, 4(3), 343-372.

Chan, H., and Faff, R. (1998). The sensitivity of Australian industry equity returns to a gold price factor. Accounting and Finance, 38(2), 223-244.

Charles M. Jones and Gautam K. (1996). Oil and the Stock Markets. The Journal of Finance, Vol. 51, No. 2 , pp. 463-491

Chen, N.F., Roll, R., and Ross, S. A. (1986). Economic forces and the stock market. Journal of Business, 59(3), 383-403.

El-Sharif, I., Brown, D., Burton, B., Nixon, B., and Russell, A. (2005). Evidence on the nature and extent of the relationship between oil prices and equity values in the UK. Energy Economics, 27(6), 819-830.

Faff, R., and Brailsford, T. (1999). Oil price risk and the Australian stock market. Journal of Energy Finance and Development, 4(1), 69-87.

Faff, R., and Chan, H. (1998). A multifactor model of gold industry stock returns: Evidence from the Australian equity market. Applied Financial Economics, 8(1), 21-28.

Groenewold, N., and Fraser, P. (1997). Share prices and macroeconomic factors. Journal of Business Finance and Accounting, 24(9-10), 1367-1383.

Hammoudeh, S., Dibooglu, S., and Aleisa, E. (2004). Relationships among U.S. Oil prices and oil industry equity indices. International Review of Economics and Finance, 13(4), 427-453.

Jones, C. M., and Kaul, G. (1996). Oil and the stock markets. Journal of Finance, 51(2), 463-491.

Jorion, P. (1990). The exchange-rate exposure of U.S. multinationals. Journal of Business, 63(3), 331-345.

McSweeney E. & Worthingto A. C. (2007). A comparative analysis of oil as a risk factor in Australian industry stock returns, 1980-2006. Global Finance Journal.

Poon, S., and Taylor, S. (1991. Macroeconomic factors and the UK stock market. Journal of Business Finance and Accounting, 18(5), 619-636.

Sadorsky, P. (1999). Oil price shocks and stock market activity. Energy Economics, 21(5), 449-469.

Sadorsky, P. (2001). Risk factors in stock returns of Canadian oil and gas companies. Energy Economics, 23(1), 17-28.

Sadorsky, P., and Henriques, I. (2001). Multifactor risk and the stock returns of Canadian paper and forest products companies. Forest Policy and Economics, 3(3-4), 199-208.