Natural Resource Information for Economic Development

Raymond F. Mikesell

• Publisher: RFF Press
• Year: 2013
• ISBN:
  9781135993092
• Edition: 1
• Topic: Biological Sciences
• Subtopic: Ecology
• Index: Biological Sciences

This book discusses the problems of developing quality information on the availability of natural resources.

Originally published in 1969

VI:
Some Guidelines for Organization and Administration of Information Activities

Our search for guidelines for the management of information activities rests on a general view about the role of information in economic decisions about natural resources. The goal is not the amassing of “complete” information, nor is it particularly advisable to move toward this goal as “rapidly as possible,” for what sensible meaning could these goals have? In the most general terms, we want information that will permit suitable investment decisions in the years immediately ahead—suitable, that is, in view of the fact that information has a real cost and that the course of economic development is powerfully constrained by forces whose operation is not going to be upset by the generation of any quantity of information whatsoever. Not only may more of one type of information mean less of others, but we run the risk of accumulating information, not for the solution of an actual problem by use of a complete information model, but for the sake of information itself, that is, for the accumulation of information that may never be used. We can not collect all the information “necessary” to solve a specified list of problems, partly because the particular problems to be solved “now” depend on information gathered earlier and partly because it pays to have only enough information to reach an acceptable solution.

How can we proceed to our rules for economizing? A conceivable method would be to emulate the agricultural economist who, together with his technical brethren, may be able to tell us how to handle the application of a certain type of fertilizer. On the basis of experiment or by statistical analysis of (varied) non-experimental fertilizer applications, he can say that with soil of a certain type, a certain pattern of rotation, and with certain climatic conditions, it will pay to apply a certain quantity of fertilizer per acre per season with present costs and prices.

For obvious reasons, no procedure very close to this is now possible with natural resource information, nor does such a procedure seem feasible on a large scale in the future. The benefits flowing from natural resource information are often quite diffuse. Not only that, but much information activity must anticipate the time of its use. And, unlike the case of the fertilizer, it is not easy to specify the product of some types of information. Superficially, some types of information may seem less than valuable because their collection serves only to cull out losing prospects—prospects, however, which could have absorbed a lot of investment funds had they been continued.

However difficult it is to make numerical estimates of the value of a little more or a little less information, two procedures are within our power. First, it is possible to make some suggestions for economizing which, while they do not result in a fine adjustment of outlays in the light of product produced, can be of help in preventing some gross misallocations of funds. Also it should be recalled that in many situations it is possible to be reasonably sure that a certain increase in expenditure will result in a product whose value is greater—or less—than the outlay in question.

The second possibility for dealing with the allocation problem involves a search for forms of organization and types of relations among organizations that will facilitate the making of judgments about marginal changes in information and its benefits. Even though we can’t estimate the value of information, we may be able to specify some of the features of an organization that will enable improved evaluation of marginal changes in the information programs as time goes along.

The suggestions that are made here have two general characteristics. First, they look toward specialization and a full use of the specialist in the execution of natural resource information programs. But the specialist must be constrained to operate within the system by which investment and operating decisions in the natural resource field are reached. He is the servant of this system.

Second, the goal should be a system in which information programs are adjusted to the needs and demands of the governmental agencies and private entities that use natural resource information. It is to be expected that the balance of the programs and their content will change as time goes on.¹

How Much Information?

Our first guideline is obvious and perhaps rather formal, but it needs to be pointed out and kept clearly in mind.

The quantity of information collected should be increased so long as the present value of the investment opportunity (or cost savings if this is the use to which the information is put) is increased by more than the cost of the information.

This problem has been discussed at considerable length in Chapter V. Here the discussion is simply a reminder of the main lines of the argument but can not very well be passed over, for the general orientation toward the management of information as a part of the process of maximizing the present social value of the investment opportunities undertaken is essential if we are to avoid uneconomic accumulation or failure to accumulate information.

Our analysis began by considering the simple case in which we suddenly become aware of an investment opportunity with constant demand and cost conditions after construction is completed. The proper size depends on one or more aspects of the behavior of a natural resource stochastic variable, for example, the flow of a stream. We can estimate the relevant characteristics of the behavior of this variable only on the basis of observed data. The problem is, how long can we afford to wait for data to accumulate before constructing?

The main problem here is the estimation of a function which relates expected present value of the investment opportunity at time t to the number of years of data available, where t is also the number of years of data available at the time of construction, t. We may think of t = 0 as “now,” the moment in which the investment opportunity is perceived.

The expected value of the investment opportunity will be low with little data available, but will rise with more data available. With little data available, the structure often² would be too large (resulting in unused capacity) or too small (resulting in lost opportunity for product or structural failure), thus reducing the expected present value of the opportunity. For small enough quantities of data, the expected value will be negative.³ As the quantity of data increases, expected value (always as of time t) will rise but at a decreasing rate. With our assumption of constant price and cost conditions, expected value will approach a maximum as our estimates of the values of the relevant parameters of the distribution involved approach the true values.⁴

We can go immediately to the conclusion that construction should take place when the cost of getting one more year of information is equal to the resulting increase in expected present value. The cost of getting one more year of data is made up of two elements, the outlay during the year to get the data, k, and interest on the expected present value of the opportunity we would have experienced if we had not waited one more year. That is, if V(t) is the basic function, we should wait until its rate of increase, V’(t), is equal to [rV(t) + k], where r is the rate of discount (the rate of return on investment).

Several conclusions are evident. First, it never will pay to wait for “complete” information. Second, an extremely important element of the problem is the cost coming from postponement of the stream of net revenues from the project. This factor means it does not pay to accumulate data until the increment in expected value is equal to the annual cost of the data. The

rate of increase per year in expected value

must not be less than

.⁵ This means the effect on expected value of one more year of data must be quite sizable to justify waiting.

Now let us change the problem to a form that corresponds better to the problem facing information agencies. Suppose we are aware of the existence of a possible investment opportunity which will become viable some day because demand is growing.

The simplest case of demand growth is very unrealistic but still instructive. Assume that demand explodes into existence full blown at a certain time. How many years ago should we have started collecting data? If we are free to choose the time for construction, it is obvious that this should be when the demand comes into existence. What we must compare is the cost of reaching back in time for one more year of data (accumulating this cost to the date of construction at the rate of discount) with the resulting increase in expected value. In this case, one more year of data does not require postponing the stream of net rents, but on the other hand we incur an interest cost, which reflects real opportunities for investment that have been given up, on the funds spent on developing information.

For more realistic cases in which demands grow from year to year, the problem is more complex. We may think of the immediately preceding question—by how many years should the data program anticipate construction—as being analyzed separately for each possible date of construction. For each of these cases there will be a maximum expected value (discounted to the date of construction) associated with a certain number of years of data. The maximum of these maxima, with all of these discounted to a common date, represents the best time for construction and will also indicate the time when the data program should be begun. It should be possible to make estimates of this type, at least retrospectively.

Quantitative estimates of the sort required by these simple models are not feasible for many problems of information. In some cases there is difficulty in quantifying the information. In others the benefits are very diffuse or hard to measure. In all cases, however, the simple model poses the relevant questions. We should be trying to judge the product that will result from more information and should bear in mind the powerful role that should be played by the rate of discount in making these judgments. Postponement of a stream of net rents is very costly, as is also the premature assembly of information. These are not “merely financial considerations” but reflect the fact that capital goods have a return in real product.

What Areas Are Suitable?

For the most part, natural resource information programs should be concentrated in or near the areas already being exploited.

Although this suggestion may seem surprising, the considerations supporting it are quite compelling if expenditure on information is to be evaluated as a public investment. There are some partial exceptions, however.

First of all, as economic product and population grow in settled areas, it will pay to increase all the inputs that are used with natural resources, and for certain types of resources one of these inputs is information about the resource. One may think, for example, of soil and information about it. In this particular case, and in some others, too, the capacity of the owners or managers of natural resources to use information grows over time. The level of education may be rising, the extension service may have become more effective, or they simply may have learned how to put the available information to work and are now ready to use more detailed and more sophisticated information. Thus the need for information about natural resources does not vanish just because the natural resources of an area are being exploited. The emphasis may change as among resources and also the type of information may change, but the demand or need will remain.

But not only do there continue to be opportunities to invest in more information in areas already under exploitation. The very fact of the present locational pattern of economic activity implies that entrepreneurs have found the obstacles to the extension of economic activities to other areas to be insuperable, given the alternative investment opportunities available and the necessity of covering costs. In parts of some countries there are areas so disfavored that economic activity in them is not only sparse but nonexistent for all practical purposes.

Natural resource survey information is not likely to diminish the impact of the forces that have resulted in some areas having little or no productive activity in the natural resource industries. The extension of economic activity to very far distant or idle areas has two strikes against it, and in some cases more.

First of all, thanks to indivisibilities that are present in the construction of many types of infrastructure, it is certain that the cost of infrastructure services will be higher in areas with a low density of economic activity. It is not possible to construct a 100-mile highway...

Table of contents

• Cover
• Title
• Copyright
• Original Title
• Original Copyright
• Contents
• Preface
• Acknowledgments
• Introduction
• I The Role of Natural Resources in Economic Development
• II The Capabilities and Limits of Information Programs
• III The Costs of Natural Resource Information
• IV The State of Natural Resource Information in Latin America
• V Scheduling Expenditures for Natural Resource Information
• VI Some Guidelines for Organization and Administration of Information Activities
• VII The Experience of Peru and Chile, and Some Problems of Small Countries
• VIII Summary
• Selected Bibliography
• Index