Essay on Industrial Real Options

Introduction

Dixit and Pindyck (1994) and Trigeorgis (1996) explored thoroughly in their comprehensive text books on real options the notions of irreversibility, the ability to delay an investment, and the option to invest, and their importance in strategic capital budgeting, which is the expanded version of the traditional discounted cash flow approach. Their conceptual framework is grounded on the theory of option pricing and financial derivatives market that are applicable towards the opportunities (or time-value options) to acquire, expand, contract, switch, suspend, and abandon real assets (i.e., real options). They developed the basic theory of irreversible investment under uncertainty and emphasized the option-like features of investment/divestment opportunities upon which the optimal decision rules are based. From this theoretical work, many aspects of the firms' investment behavior deserve empirical verification whether or not they conform with its hypothesis.

One aspect that interests us from the standpoint of industrial organization is the determinant of the firms' investment behavior itself. In a highly deregulated market economy such as of the U.S., firms operate and compete freely with less government's structural interventions, even though certain kinds of discretionary fiscal and monetary interventions are implemented to achieve domestic economic goals and to prevent market failures. Nonetheless, their products and services are geared toward two distinct types of demand; one is the demand for public goods and the other for the more specific private goods. Since the demand for public goods is less sensitive or inelastic to price changes than the demand for private goods because of their necessity and high predictability characteristics, income derived from the former type of demand tends to be more stable than from the latter.

Given that perfect competition in the supply of both kinds of goods holds in the U.S. market, public-goods firms are considered relatively equally as efficient and competitive as their counterparts in the private-goods firms. Thus, we can comfortably imply from this presumption that the expected income or future cash flows of the public-goods firms are less volatile than those of the private-goods firms.The establishment of the perfectly competitive market environment in both kinds of goods in the U.S., with the only exception in demand elasticity difference between them, enables us to further imply that the optimal investment decision for the public-goods firms is to make a larger amount of investment commitment in their fixed assets relative to the private-goods firms. This assertion is therefore the main hypothesis to be tested in this study.

Review of Real Options Literature Back to Top

There has been no published literature that encompasses the aspect of real options implication on public-sector industries. Apart from the two main text books by Dixit & Pindyck and Trigeorgis mentioned earlier, there are some research papers on real options by themselves and other authors that address investment/divestment behavioral models in a general firm's context. Dixit (1989) examined the firm's entry and exit decisions when the output price follows a random walk, and the importance of hysteresis effects on those decisions. Pindyck (1993) theoretically discussed and modeled the irreversible investment decisions when projects take time to complete and are subject to uncertain costs. Trigeorgis (1990) demonstrated the application of an option-based numerical analysis method in the case of a natural resource investment opportunity. His subsequent papers focused on a generalized model using a log-transformed binomial numerical analysis method to value a complex multi-option investment (1991), and a nature of option interactions and the valuation of capital budgeting projects with flexibility in the form of multiple real options (1993).

In the area of temporary shut down or mothball option, McDonald and Siegel (1985) explored and analyzed a discrete project in which operation can be suspended when profits are negative and resumed at no additional cost if they subsequently turn positive. Brennan and Schwartz (1985) utilized the convenience yield derived from futures and spot prices of a commodity to determine the value of the options to temporary shut down and reopen a mine and to abandon it for salvage, but do not address the interactions among individual option values. Myers and Majd (1990) analyzed the option to abandon an asset for salvage value in relation to the asset's life. Merville and Mishra (1991) approached the real options from the right-hand side of the balance sheet discussing the relationship of capital investment and firm leverage. Grenadeir (1995) used real options approach to value lease contracts of several forms.

Yet, the relationship between real options and the nature of the firms' products and services has not been formally surveyed. Once this relationship has been established, it could lead to many other research interests in the fields of political economy, game-theoretic and information economics, as well as international finance, capital mobility, and foreign direct investment (FDI). In terms of domestic political economy implications, valuation of real options for various public expenditure programs would enhance the awareness of the government in designing appropriate investment incentives and formulating the flexible yet credible investment policies to stimulate or stabilize the economy. By intervening to alter real-options value of the investment through the manipulation of information asymmetry and policy discrimination in certain economic sectors, the government can signal ex ante or screen ex post the market participants to alleviate the problems of moral hazard and adverse selection which are the major causes of market failure. For international finance and capital movement issues, the less developed countries' governments can use the real-options framework to increase the attractiveness of state-owned enterprises' privatization programs for high-quality FDI and equity-based foreign capital inflows which not only bring in long-term investments but also transfer desirable technologies to the host countries. Research questions in these areas can be investigated in the more quantitative terms which will lead to many other theoretical and empirical issues of those interrelationships between the public-sector and the private-sector economies both domestically and internationally.

Research Methodology Back to Top

This study can be perceived as being a combination of exploratory and descriptive researches. It employs certain research techniques that would allow us to tackle both the breadth and the depth of the available data while providing flexibility from devising some normative criteria. This methodology section is divided into three parts: 1) research design approach, 2) sample design and data collection, and 3) data analysis and interpretation.

Research Design Approach

The objective of this study is to explore the evidence of real options from the secondary data available from Standard & Poors' CompuStat database through the comparison between two groups of sector-oriented industries - public-goods and private-goods firms - which are our main sources of variation (i.e., explanatory variable). The criterion or dependent variable on which its variation would be measured is the ratio between the industries' gross fixed assets (including property, plant, and equipment) and their total assets. The use of gross fixed assets instead of net fixed assets helps us to avoid the confounding from other source of variation generated by the different methods of depreciation. The approach for descriptive statistics is to conduct both cross-sectional and time-series (or longitudinal) analyses in a sequential order - cross-sectional first then pooled five-year time-series second.

Our research design's steps are as follows. First, the data will be classified into two main sectors labeled 1) public-goods and 2) private-goods, under which each has five accompanying subgroups being classified according to the SIC numerical index. Second, the data on fixed assets and total assets are manipulated cross-sectionally using total assets-weighted average method for each year from 1990 to 1994. Third, the averaged fixed assets to total assets ratios are pooled time-serially to account for any variation, if any, due to changes in time. And fourth, the results are tested to see if there are any statistically significant differences between the two groups' means, which will enable us to conclude that real options are more empirically evident in the private-sector than in the public-sector industries.

Sample Design and Data Collection

There are currently 9,469 firms listed in the S&P CompuStat database including the foreign firms whose stocks are American depository receipt (ADR) issues. These firms are grouped according to the standard industrial classification (SIC) indices. The financial data of firms with the same SIC number are aggregated to comprise the industry's financial data. Based on the two-digit SIC numerical index, 12 industries supply public-goods to the market.

SIC	Industry
40	Railroad Transportation
41	Transit and Passenger Transportation
42	Motor Freight Transportation
44	Water Transportation
45	Air Transportation
46	Pipeline, except Natural Gas
47	Transportation Services
48	Communications (Telephone, Radio, TV, Cable TV)
49	Electric, Gas, Water, and Sanitary Services
80	Health Services
82	Educational Services
83	Social Services

For the comparative purpose of this study, these public-goods industries are regrouped and handpicked in such a way that they can be matched as closely as possible with their counterparts in terms of complementary goods and services in the private-goods industries. However, educational and social services do not have any representative firms listed in the database. The regrouping occurs in the transportation industries where they are clustered together as one new big group, including railroad, transit and passenger, water, and air transportation industries but excluding motor freight, pipelines, and transportation services industries. Another industry that is subject to regrouping is the electric, gas, water, and sanitary services. The first three industries, i.e., electric, gas, and water, are clustered together to form the utilities group, whereas sanitary services are singled out to be its own separate group. As a result of regroupings, our candidate public-goods sector will be composed of:

With the total number of 788 firms comprising the public-goods sector, the same number for the private-goods sector is sought in order to balance them and to ensure that they are operating within a similar competitive environment, i.e., with the same number of competing firms within each industry and across the two sectors. The four-digit SIC index provides a more detailed information about the private-goods industries in terms of types and number of firms which allow us to select the most appropriate and complementary industries for matching and balancing with those in the public-goods industries. The regroupings and handpickings of the private-goods sector are given below:

The source of data from which we use to derive and manipulate the required variables is the aggregate annual balance sheet of each industry from 1990 to 1994. The ratios of gross fixed assets to total assets are calculated and then tabulated in Table 1. Table 2 shows the industry groups' mean ratios are derived cross-sectionally from the total assets-weighted average for each year. These proportional weights are given in Table 3. After the cross-sectional mean ratios are obtained, the time-series data are pooled and regressed to arrive at the general linear model statistical results within the ANOVA framework as given in Table 4.

Table 1
Gross Fixed Assets to Total Assets Ratios

1)Transportation:

2) Transportation Equipment:

3) Communications:

4) Communication Equipment:

5) Utilities:

6) Electrical Equipment:

7) Sanitary Services:

8) Food Products:

9) Health Services:

10) Pharmaceuticals:

Table 2
Cross-sectional Total Assets-Weighted Average GFA Ratios

Group of Industries	1990	1991	1992	1993	1994
Transportation	1.837	1.929	2.003	1.999	2.085
Transport Eqm.	0.449	0.474	0.495	0.511	0.525
Communications	1.219	1.168	1.152	1.168	1.183
Comm. Eqm.	0.409	0.389	0.402	0.392	0.392
Utilities	1.110	1.152	1.139	1.092	1.116
Electrical Eqm.	0.340	0.364	0.352	0.322	0.388
Sanitary Services	0.692	0.713	0.738	0.754	0.752
Foods Products	0.477	0.487	0.516	0.532	0.543
Health Services	0.696	0.673	0.663	0.700	0.608
Pharmaceuticals	0.523	0.527	0.539	0.552	0.509

Table 3
Proportional Weights Based on Total Assets

1) Transportation:
2) Transportation Equipment:
3) Communications:
4) Communication Equipment:
5) Utilities:
6) Electrical Equipment:
7) Sanitary Services:
8) Food Products:
9) Health Services:
10) Pharmaceuticals:

Data Analysis and Results Interpretation

The following regression model is used in our data analysis:

Y_ijt = a_t + b_iI_it + g_jS_jt+ e_ijt
where: Y_ijt= Assets-weighted average gross fixed assets to assets ratio (GFA) of the i^th industry group in j^th sector in t^th year.; a_t = Based-level GFA ratio in t^th year.; b_i= Industry parameter.; I_it= Industry dummy variable in t^th year.; g_j = Sector parameter.; S_jt = Sector dummy variable in t^th year.; e_ijt = Random residual.

The corresponding effect model for these same data for ANOVA is as follow:

m_ijt = m + a_t + I_i + S_j
where: m_ijt = Combined mean effects; m = Overall mean effect; a_t = Longitudinal mean effect (Time-series treatment variable); t = {1990, 1991, 1992, 1993, 1994}; I_i = Industrial mean effect (Cross-sectional treatment variable); i = {transportation vs. transport equipment., communication vs. communication equipment, utility vs. electrical
equipment., sanitary vs. foods, and healthcares vs. pharmaceuticals}; S_j = Sectoral mean effect (Blocking variable); j = {public sector, private sector}

The results of general linear model regressed under the ANOVA framework analyzed using the SAS statistical package are shown in Table 4 below.

Table 4 General Linear Regression & ANOVA Outputs

Using Fisher's Least Significant Difference (LSD) Test of different means of each source of variation, the results show that there is no significant difference in longitudinal mean effect (a_t) and industrial mean effect (I_i) except for transportation industry, but there is a significant difference between sectoral effect (S_j) as follow:

S₁ = Sectoral mean of five public-goods industry groups; S₂ = Sectoral mean of five private-goods industry groups
: H_o: S₁ = S₂ against H_a: S₁ ¹ S₂; Critical t_0.025,40 = 2.021 against LSD's t = 15.554; Decision Rule: Reject H_o if |t| > LSD t-value.

Results of the Study

From the results of statistical comparison and inference conducted in the previous section, there is a strong indication that real options are more evident in the private sector than in the public sector in its relatively lower commitment in fixed assets across industries and over time. It is expected that these results are more sharply contrasted between regulated and unregulated industries in which competitive intensity between the two groups differs.

Conclusion and Recommendation Back to Top

It is concluded that in the highly competitive markets in both public and private sectors of the U.S. economy, there exist real options such that the private-sector industries with equal number of firms and whose products and services are complementary to those of the public-sector industries will make relatively less commitment in fixed assets investment. In the same token, the public-sector industries whose expected future cash flow streams are less uncertain and less elastic to price changes than those of the private-sector industries tend to be highly committed to the long-term illiquid capital investments.

For future research in this area, it is recommended that efforts be put into the quantification and valuation of public-sector real options in different competitive environments as well as across the nations. To this research direction will the applications of real options be of greater benefits in the areas of domestic economic policy-making, dynamic game theory with asymmetric information for the prevention and alleviation of market failures, and strategic foreign economic policy initiatives of the emerging and transitional economies to attract and retain high-quality FDI. More specific research can be done at the firm level where causalities between product prices, investment costs, project's time frame, and the firm's investment/divestment behavior, as theoretically modeled in the recent literatures, can be tested.

References

Brennan, M., and Schwartz, E. (1985). "Evaluating Natural Resource Investments." Journal of Business, 58, 135-157.

Dixit, A. (1989). "Entry and Exit Decisions under Uncertainty." Journal of Political Economy, 97, 3, 620-638.

Dixit, A., and Pindyck, R. (1994). Investment under Uncertainty. Princeton: Princeton University Press.

Grenadier, S. (1995). "Valuing Lease Contracts: A Real-Options Approach." Journal of Financial Economics, 38, 297-331.

McDonald, R., and Siegal, D. (1984). "Option Pricing when the Underlying Asset Earns a Below-Equilibrium Rate of Return: A Note." Journal of Finance, 39, 261-265.

Merville, L., and Mishra, C. (1991). "Capital Investment and Firm Leverage: A Corporate Real-Options Approach." Research in Finance, 9, 49-73.

Myers, S., and Majd, S. (1990). "Abandonment Value and Project Life." Advances in Futures and Options Research,4, 1-21.

Pindyck, R. (1993). "Investment of Uncertain Cost." Journal of Financial Economics, 34, 53-76.

Trigeorgis, L. (1990). "A Real-Options Application in Natural-Resource Investments." Advances in Futures and Options Research,4, 153-164.

Trigeorgis, L. (1991). "A Log-Transformed Binomial Numerical Analysis Method for Valuing Complex Multi-option Investments." Journal of Financial and Quantitative Analysis, 26,3, 309-326.

Trigeorgis, L. (1993). "The Nature of Option Interactions and the Valuation of Investments with Multiple Real Options." Journal of Financial and Quantitative Analysis, 28, 1, 1-20.

Trigeorgis, L. (1996). Real Options: Managerial Flexibility and Strategy in Resource Allocation. Cambridge: The MIT Press.

* Worapot Ongkrutaraksa is a lecturer in Finance and Strategic Management at Maejo University's Faculty of Agricultural Business, Chiang Mai, Thailand. He used to conduct his post-graduate research in financial economics at Kent State University and international political economy at Harvard University through the Fulbright sponsorship between 1995 and 1998.

E-mail: [email protected]

Home Page