Monthly Archives: March 2005

45 – Thinking like an economist 13: What is the cost?

Economists treat costs somewhat differently to the way that accountants do. This article explains why and how.

You run a business producing, say, furniture, or maybe wheat. What does it cost you to produce a unit of output? It is a question of some importance to a prudent business manager.

Perhaps it sounds simple. Couldn’t you just add up all of your expenditures and divide the total by the number of units of output? That would give you a measure of the average cost, but not necessarily the most useful or valid one.

Economists tend to look at things through a decision making lens, and as a business manager, one of your key decisions is how much output to produce. Curly questions about costs that you need to address to make this decision include:

(a) How should you deal with the fact that some costs are fixed while others vary with the level of output?

(b) Are input costs the only relevant costs?

(c) What if you produce more than one product? How should you apportion costs when an input is used to jointly produce two or more outputs? (e.g. crops and livestock from the same farm paddock within a year).

Economists have answers, and they are a bit different from the answers you would get from an accountant. The economists’ answers are tailored to making decisions that maximise the overall benefits.

(a) Fixed costs are almost (but not quite) irrelevant to decisions about production levels. Economists focus on “marginal costs”: the cost of making one additional unit of production, or in other words, the variable costs at the margin. Because fixed costs don’t vary with the production level, they don’t appear in the calculation of marginal costs. Marginal cost generally varies unit by unit as production changes, and the manager needs to know about that (so that he or she can compare it to marginal benefits) to choose the best level of production.

(b) Input costs are not the the whole story. Economists use a much broader concept called the opportunity cost. The opportunity cost does includes input costs: the financial cost of acquiring and transforming the necessary resources into a unit of production. But it also includes the cost of not using the resources in their best alternative use. If you didn’t use your land to grow wheat, the best alternative land use might be barley. The barley profits you have to give up to grow wheat are, in a real sense, part of the cost of growing wheat. The opportunity cost, that is. Barley is the opportunity. Not grasping that opportunity results in a cost, in the sense of missed profits.

(c) You don’t have to make arbitrary divisions of costs amongst joint products. You just need to watch overall benefits and overall costs. In practice, as the quantity of any one product is increased, the most profitable quantities of production of other products will vary too (probably down, but not necessarily). The opportunity costs (or benefits) of these other changes have to be factored in during the calculation of opportunity cost for the current product, and they already account for any changes in input costs associated with the other product(s). You don’t actually have to allocate costs among the various outputs to make the best decision. By focusing on the marginal opportunity cost for any one , the problem of allocating costs among outputs becomes irrelevant.

In this framework, you can only look at the cost of one product at a time. If a firm produces several products, the marginal opportunity cost of producing any one of them will already factor in the benefits and costs of producing any or all of the others.

Obviously, then, there is no easy conversion from accounting measures of costs to economic costs. However, if you are working with an accounting style model (e.g. a budgeting model) it is possible account for the above curly issues if you approach it properly.

(a) You can look at average (or total) costs and benefits, but only if you recalculate the averages (or totals) for each possible level of output, and choose the one with the highest net benefit.

(b) You can’t easily calculate opportunity costs in a budgeting model, but you can achieve the same effect by comparing results for all possible strategies or options. All you are really doing when you calculate opportunity costs is comparing the current option with the best alternative. Compare net benefits for all options, and you have it.

(c) When you are doing (a) and (b), focus on total profits for the whole business enterprise, rather than trying to calculate average costs for individual outputs.

Do all that and you’ll be like a human economic optimising machine. In practice, computer-based optimisation models of firms or industries built by economists often do exactly that. They are structured somewhat like an accounting model, but are set up to process the data and produce results in a way that is consistent with the sort of economic model you see in an economics text book.

David Pannell, The University of Western Australia

44 – Thinking like an economist 12: Hand it over!

The demand curve is a deceptively simple concept that economists use constantly. It has at least three different valid interpretations that make it useful in different roles.

“Now look! Give me a copy of the new Elvis Costello CD, right now. Hand it over!”

That’s a demand.

“I’d like to buy a copy of the new Elvis Costello CD please.”

According to economics, that constitutes demand too.

“Demand” is one of a number of words that have been given somewhat different meanings in economics than in common English. (Others that spring to mind include “public”, “surplus” and “cost”.)

Demand in economics does not carry any connotation of authority, rudeness or urgency. It simply refers to a consumer’s voluntary choice to purchase some quantity of a good. It’s a deceptively simple concept that economists use constantly.

How much would they choose to purchase? The answer to that question is their level of demand. It might be communicated in a most undemanding way. It might not even be communicated at all, in the case of some public goods for which free riding is possible (e.g. some environmental goods – see PD#22). But such goods still face demand from those who benefit from their consumption, albeit hidden demand.

Demand for a product can be affected by a number of variables, including characteristics of the consumers (their income, tastes, preferences), the availability and prices of other products that are seen as substitutes, and the price of the product itself. Economists being economists, we tend to take tastes and preferences for granted and focus on prices and, to a lesser extent, incomes. So much so that when we talk about a “demand curve” or a “demand function” we are usually referring to the relationship between demand and the product’s own price.

Everyone knows that when the price of a product rises, consumers choose to consume less of it.

Or do they? There can be some interesting exceptions to this rule. According to Bill Bryson in Made in America, “When Kentucky Fried Chicken introduced ‘Extra Crispy’ chicken to sell alongside its ‘Original’ chicken and sold it at the same price, sales were disappointing. But when its advertising agency persuaded it to promote ‘Extra Crispy’ as a premium brand and to put the price up, sales soared.”

An example close to my interests: in the late 1970s there was an Australian maker of high quality guitar amplifiers who tried to compete with established brands like Fender and Marshall by being very efficient and offering very low prices. They struggled on for a while with poor sales, before deciding that consumers were using price as an indicator of quality. They doubled their prices, and sales went up nicely.

But these price changes are confounded with marketing. Normally, higher price, lower demand. There are thousands of statistical studies that bear this out.

There are two main causal factors for this, both common sense. One is that as the price of a particular good goes up you can afford less of it (and of all goods, for that matter). The other is that at a higher price, the good is less attractive as a use for your hard earned cash than other things start to look.

For some goods, the quantity demanded drops away rapidly as price rises, while for others we are almost impervious to price changes. This reflects the strength of our psychological need or preference for the good, which depends in part on the availability of good substitute products. Even if we have a pretty fixed demand for fresh fish overall, demand for one species might fall rapidly as its price rises above that for another similar fish. They are good substitutes, making their individual demands more sensitive to price.

You might consider that economists are inordinately obsessed with demand curves. Perhaps we are obsessed, but it’s for good reasons. They are extremely useful and informative, especially when we turn them around and treat quantity as a function of price. That is, we ask the question, “If the quantity of a good available for consumption was limited to X, what would the price have to be to result in consumption of exactly X?” Neatly, you can read that off the demand curve derived by asking quite a different question.

The value of thinking about the relationship in this “inverse” way is two fold:

(a) it allows us to examine the otherwise complex interaction between demand and supply in a very simple way, including predicting what is likely to happen to price if either demand or supply shifts. You’ve all seen supply and demand curves overlaid.

(b) it provides information about the monetary value that consumers place on consumption of the good. This is useful for purposes such as benefit-cost analysis.

On the other hand thinking about quantity of demand as a function of price (i.e. the right way around) allows us to examine price as a tool for manipulating behaviour (e.g. see PD#41). Overall, it is suprising how many different ways there are to think about and use such an apparently simple relationship.

David Pannell, The University of Western Australia

43 – New knowledge or new technology: which is better?

Other things being equal, R&D to develop a successful new technology will very often be worth more than R&D that successfully generates improved information for decision making about the same issue. Getting wrong the distinction between these two types of research, or failing to make the distinction, can make a big difference to the estimated value of research.

What is the value of research? It is a complex and multi-facted problem. Thinking specifically about agricultural research, the answer depends on a long list of factors, including:

(a) The predicted or estimated biological, technical and/or management changes from implementing research outcomes.

(b) Any negative or positive side effects (internal or external to the farm) resulting from implementation of the research. This would include any environmental effects and price impacts from changes in supply or demand.

(c) Costs to the farm firm of implementing findings from the research.

(d) Given (a), (b), and (c), the potential economic benefits per hectare or per farm (net of costs to the farm firm but not of research costs).

(e) The scale of potential benefits: the number of hectares or farms potentially affected.

(f) The proportion of the potential scale for which adoption occurs, and the timing of the adoption.

(g) The probabilities of different levels of success from the research.

(h) Direct costs of undertaking the research over time.

(i) The discount rate.

(I am talking about applied research here. Valuing basic research is more difficult again.)

From this long list, in my experience the items with the biggest influences on the value of research are (d), (e), and (f). When looking at attempts to value research benefits (d), a common error is to fail to distinguish correctly between research that generates information (for improved decision making) and research that generates a new technology. Other things being equal (i.e., the other things on the above list), a successful new technology will very often be worth more than successfully applied new information about the same issue. Getting the distinction wrong, or failing to make the distinction, can result in very wrong estimates of research value.

Why is that? First note that payoff functions from management decisions in agriculture often have a relatively flat payoff curve near the maximum. For example, the two payoff curves in the figure below have relatively large ranges over which changing an input level makes very little difference to payoffs. In other words, research that provided information to help fine tune the input level will make little difference to payoffs. As long as you use an input level that gets you reasonably close to the top of the hill, going precisely to the top of the hill doesn’t get you might higher.

To illustrate, suppose that without research, the perceived payoff to different input levels follows the lower curve. Suppose now that new research provides information that the true payoff curve is actually the higher curve. You thought it was one thing – it turns out to be something else. What is the value of this information? Without research, the input level that appears to be optimal is I1 (i.e., that is the input level that appears to give the maximum possible profit) and the anticipated payoff from this input level is P1,1. After the research, the optimal input level is revealed to be I2 and the anticipated payoff from this input level is P2,2. With the new information, the input level corresponding to the top of the hill appears to have shifted. However, the research does not actually shift the payoff function, it only provides improved information about where it is (and where it was all along). If the higher payoff function is actually the true one, then application of the original input level I1 would have resulted in payoff P1,2. The improved payoff resulting from the research is the vertical distance between P1,2 and P2,2 – not a big value.

On the other hand, the outcome of research to develop new technologies is more likely to be an actual rise in the payoff function (e.g. through breeding of higher yielding crop varieties). If the shift in the payoff function illustrated in the figure above was an actual shift resulting from R&D (rather than a perceived shift as previously), the value of the R&D to this decision maker would be the full difference P2,2 – P1,1. The benefits of this type of R&D are not reduced by the presence of any flat payoff function. From this it can be seen that, other things being equal, there are good reasons to expect that successful R&D intended to improve management decisions about existing technologies will pay off less than successful R&D to develop new, higher performing technologies.

For similar reasons, payoffs from agricultural scientific communication (extension or technology transfer) that provides information useful for management decisions about existing farming technologies will also struggle to overcome the influence of flat payoff functions. Unless the information suggests management practices that are substantially different to those in current use, increases in payoffs are unlikely to be large. This suggests that extension providers should target issues where the technologies are new and unfamiliar to farmers, or where for some reason farmers have developed clear and important misperceptions about the technologies.

David Pannell, The University of Western Australia

Further Reading

Pannell, D.J. (2006). Flat-earth economics: The far-reaching consequences of flat payoff functions in economic decision making, Review of Agricultural Economics 28(4), 553-566. Prepublication version here (44K).

Pannell, D.J. (1999). On the estimation of on-farm benefits of agricultural research, Agricultural Systems 61(2): 123-134. full paper (61 K)

42 – Research meets policy

This week’s PD is just an advertisement for a useful little booklet produced by Land and Water Australia, called “Research Meets Policy: improving the uptake of your research”. It outlines issues and strategies for researchers who wish to connect with policy makers to have their research make a difference. It is available for download here.

David Pannell, The University of Western Australia

Further Reading

Pannell, D.J. (2004). Effectively communicating economics to policy makers. Australian Journal of Agricultural and Resource Economics 48(3): 535-555. full paper from journal (138K pdf) also available via the Journal homepage: