2.2 Economic Decision-Making
Businesses, such as General Motors, IBM or Time-Warner, are able to earn money for their shareholders by receiving payments from their customers that exceed payments made to their suppliers. Their ability to do this depends critically on the presence of ďmoney on the tableĒ. That is, customers will not pay more for a product than the benefits they derive from it, and suppliers will not accept payments that do not cover their own costs. So ultimately, for there to be something left for the business, a customerís actual benefits must exceed the suppliersí actual costs.
By bringing customers and suppliers together, businesses can create value, and† can ensure that they appropriate some of this value for themselves as profit. For this reason, the first important set of tools of economic analysis is concerned with identifying value and its sources so as to understand the role of a business in value creation.
Value is not often a readily quantifiable concept. It cannot simply be reduced to monetary terms. A natural question to ask is how profit, which is a distinctly monetary measure, can arise from value, with elements that are often not monetary? The answer lies in the way in which money, something people and businesses prefer to have more than less of, can assist in guiding the decisions of rational agents.
The benefits a person derives from consuming an ice cream cannot be readily quantified; however, that same person can be asked to name the highest price that they would be willing to pay for an ice cream. This would give a monetary equivalent for the benefit that that person places on ice cream. Moreover, it can be related back to the payment that an ice-cream supplier would need to receive in order to cover supply costs.
By stepping into the shoes of key decision-makers, you can potentially determine the monetary equivalents of different actions. This exercise allows you to analyse whether there is an opportunity to create value. In this topic, you will learn how to put yourself in the place of decision-makers to establish the existence of value-creating opportunities for business.
We begin our study of decision-making with the simple case of non-strategic decisions. A decision is strategic if it requires you to take into account how others will react to your decision. For instance, if your firm raises the price of its product, it has to consider how your competitors may react. Strategic decisions are the focus of segment 5. In a non-strategic environment, on the other hand, either there are no other competitors or their response can be predicted. Tools such as decision analysis are aimed at decision-making in the non-strategic environment. Some of the concepts presented in this chapter will therefore be familiar to those who have a background in decision analysis. The advantage of economics, and more specifically of Game Theory, is the ability to expand decision-making into strategic environments.
A common way of representing decisions is to use a tree-like structure called a decision tree. Decision trees are made up of nodes and branches, which are used to represent the sequence of moves and the actions, respectively.
Here is a decision tree for a landlord who is choosing between leasing a property or using it as a base for their own business.
In this subject, decisions are represented by square nodes. Node L is the decision node where the landlord chooses between leasing or using the property themselves. Node L is the first node and is known as the initial node. The triangle-shaped ending nodes on the right are the terminal nodes, which also have the payoffs to L associated with each outcome listed beside them.
Notice that the landlord earns more by leasing the property ($120,000) than by using it for their own business ($100,000). Consequently, the landlord should choose to lease the property.
In addition to decision nodes, a decision tree can include chance nodes. These are represented in this subject by circles.
To see this, suppose that each of the choices the landlord faces involves some risk. The tenant may become bankrupt and will be unable to paythe rent. The landlordís own business might not be as profitable as was forecasted. Decision trees can incorporate this type of uncertainty over payoffs.
This decision tree incorporates uncertainty over the returns to leasing or using the property for the landlordís own business. If the landlord leases, there is a 20 percent probability that the tenant might go bankrupt and not pay rent. This is depicted by the 0.8 and 0.2 numbers on each of the branches emanating from the chance node. If the landlord uses the property for their own business, there is an equal chance it will earn profits of $100,000 or $80,000. This is reflected by the 0.5 numbers on each branch.
In order to evaluate the lease decision, the landlord needs to look forward and work backwards. This involves first calculating the expected payoff from each decision. For the lease, this is (0.8 x 120,000) + (0.2 x 0) = 96,000. For the own business, this is (0.5 x 100,000) + (0.5 x 80,000) = 90,000. Taking this into account reduces the decision tree back to a tree when there is no uncertainty.
Thus, taking into account the uncertainty over payoffs, we can reduce the landlordís problem to a choice between leasing and obtaining an expected profit of $96,000 and using it for her own business and obtaining an expected profit of $90,000. Faced with these choices, the landlord should still choose to lease.
Click here to read about the costs associated with risk.
When deciding whether to undertake an advanced degree such as an MBA, students need to make similar judgements regarding options available and the likely outcomes to their income and career if they obtain an MBA or not. Before you see an example of how one prospective student made her career decision, click here to review the concept of net present value (NPV).
Now, to see how the prospective student made her career decision, click here.
Decision trees are useful in that they can assist in identifying some common pitfalls in business decision-making. Here we illustrate four of these.
Costs that have already been incurred and cannot be recovered are known as sunk costs. When making decisions, managers should ignore these costs; otherwise, they risk making poor decisions. The following example demonstrates this point:
A Texan oil company is considering investing US$40 million in new leases of land in the Gulf of Mexico for oil drilling. Its geologists favour the plan, saying that the company has already spent US$200 million on oil exploration in that area and that the company needs to see the project through or the money will have been wasted.
The company's chief financial officer tells the geologists that their logic is flawed. He points out that because the company spent the money to acquire the exploration information, there is no way to recover it. He demonstrates this logic by constructing the relevant decision tree.
If the company (C) spends $40 million on new leases and has an expected return from this of R, its total payoff from this activity is R Ė 40 million Ė 200 million. If it does not invest, it still faces a loss of $200 million. Comparing these two activities, the company chooses to invest in the lease if R Ė 40 million Ė 200 million > -200 million or R > 40 million. Thus, the decision is independent of the total amount spent on exploration to date.
The money spent on previous exploration is a sunk cost. Whether the oil company continues to drill in the area or uses the $40 million for another project, the $200 million they have already invested cannot be recovered. Using this sunk cost to justify continued investment could lead to greater losses. Investment should only continue if the expected return exceeds the investment costs from this time onwards ($40 million).
This suggests a useful principle.
When making decisions, managers need to be forward-looking and thus should ignore sunk costs.
To learn more about sunk costs, view the following animation.
Firms usually are making several decisions at the same time. It is critical, therefore, that a decision whose outcome is independent of another decision is recognised as such and not driven by those decisions.
To see this, letís return to our oil-drilling example. Suppose that the refining operation of the oil company leaps into the debate and argues that new leases should not be purchased because the company needs to make an investment of US$50 million in upgrading the efficiency of its existing refining operations.
In this instance, the chief financial officer sides with the engineers in the debate. He points out that the decision tree looks like the following (where we ignore past exploration costs because they are sunk):
Here S is the level of cost savings that is expected to be achieved by the plant upgrade. In choosing whether to upgrade or not, if a new lease has been purchased, the company compares payoffs of R - 40m + S - 50m to R - 40m and will upgrade if S > 50m. On the other hand, if the new lease is not purchased, the company compares S - 50m to 0 and upgrades if S > 50m. Regardless of the decision on new leases, the decision to upgrade or not rests on the same factors. That is, the company should upgrade if S > 50m regardless of whether it purchases new leases or not. The upgrade decision is, therefore, irrelevant to the decision as to whether to purchase the new leases or not and should not be factored into the debate.
This suggests a useful principle:
When making decisions, ignore other decisions: the outcome of which does not impact upon the decision at hand.
Care must be taken, however, in applying this principle. You need to work out if a decision is truly independent. For example, if the refining plant was considering an expansion in capacity, it may be relevant whether new oil reserves were available or not. In this case, the decisions to upgrade and continue exploration would be interdependent.
When considering increasing or decreasing their production levels, firms should base their decisions on marginal cost (MC) rather than average cost. MC accurately reflects the cost of producing an additional unit of output or the savings from producing one unit less. To see how mistakes can be made by not considering marginal costs, consider the following:
A railway owner notices that their engines have the capacity to haul more cars and is considering putting on an additional car. At present, they receive $1,000 per carload and operate 10 of these and this would not change if an 11th car was added. A look at the accounts of the company confirms that the cost per carload consists of $600 for the rental value of the car, $300 for the allocated cost of the rail (that is, $3,300 divided by 11) and $200 for physical loading and unloading. Therefore the average cost of each carload is 600 + 300 + 200 = $1,100 per load.
If one were to compare this to the price per carload, one might be tempted to consider an expansion unprofitable. However, framing the decision in a tree reveals a different picture.
Notice that the company makes a loss regardless of whether it uses 10 or 11 cars. However, comparing those losses indicates that the company makes $100 more if it adds a car than if it does not. This happens even though on average carloads are unprofitable.
A closer examination reveals what is going on here. Notice that adding another car only adds car rental and loading costs, ie, MC is $800, but does not change the rail cost. The rail costs are fixed cost that would be unchanged if more cars were hauled. Hence, they are not part of the relevant decision here to add another car or not.
Click here to see another example of how margins should be considered rather than averages.
Average cost plays a role in some analyses, but in most cases, it is useful to follow this rule:
The proper cost to consider when trying to maximise profits is the firm's MC.
When economists talk about a good's cost, they refer to its opportunity cost, a measure not only of explicit costs like labour and materials used in production, ie, the items that are included in a firm's income statement, but also of implicit costs, such as revenues that the inputs could have generated through some other use. Consider the following example:
For the past year, a couple has owned a speciality bookstore in a major metropolitan area. The bookstore is operated in a space of a building the couple owns. In that year, the store had revenues of $559,135. One of them has calculated that total cost, including labour, the wholesale costs of books, equipment costs and marketing costs, are $431,296. He reports to his partner that the bookstore is very successful, with profits of $127,839 for the year.
His partner is unconvinced. She notes that they did not take into account the rent that the two could have earned on the space they used, as much as $8,000 per month. They also did not consider the wages of $54,000 per year he gave up by choosing to manage the bookstore. The sum of these two implicit costs, $150,000, needs to be added to the original estimate of the bookstore's costs to get the total economic cost.
To see another example of the difference between accounting and economic costs, click here
Because the implicit costs together are greater than the original estimate of the bookstore's profits, in an economic sense, the bookstore is losing money.
The bookstore example suggests an important principle:
When making decisions, economic agents should take into account the opportunity costs of actions, not merely their explicit costs.
The key point here is that when contemplating a decision, it is important to use only relevant costs in weighing up alternative options. If additional costs are brought in, then profitable decisions may be mistakenly held to be unprofitable and value-creating opportunities would be lost.
Another cost related to accounting costs is historical costs. Click here to see if historical costs should be used.
Click here to see how opportunity costs can be used in a merger.
Two critical decisions in economics are a customerís decision to purchase a good or service and a supplierís decision to produce that good or service. Before considering the value customers and suppliers might generate from trade, it is useful to consider these two decisions in more detail.
A monetary measure of the desire customers have for a product is their willingness-to-pay. A customerís willingness-to-pay is the maximum price he or she would pay and still choose to purchase a product.
This can be easily depicted in a decision tree. Suppose an antique dealer sees a Ming vase for sale in China. There will be $8,000 of costs associated with bringing the vase to his home country and selling it. The trader has a certain buyer he knows will purchase the vase for $200,000. The price the vase sells for in China is p. The dealerís decision tree is as follows:
Notice that the dealer (D) will choose to buy the vase if $192,000 Ė p > $0 or p < $192,000. Thus, so long as the price of the vase is less than $192,000, the dealer will purchase it. In this example, $192,000 represents the dealerís willingness-to-pay for the vase.
In general, a customerís willingness-to-pay for a product is determined by many factors. These include the subjective benefit or utility a customer may derive from a productís consumption, eg, a vacation package, or the increased profits that result from utilisation of a product, eg, a more modern production technology. Consumers may be willing to pay more for a product when their incomes rise, when a complementary product becomes cheaper, such as CDs and lower priced CD players, or when they are located closer to a particular firm or that firm provides a more favourable brand image. All of these factors -- often the domain of marketing subjects -- are involved in forming a customerís willingness-to-pay.
Willingness-to-pay, however, is not an absolute notion. It often depends on the alternative outcome customers face if they choose not to purchase a particular product. Thus, you need to examine the overall decision customers are facing as they evaluate purchasing and not purchasing the product.
For instance, suppose that the dealer can only carry one item back from China and has identified another piece that could be shipped. The dealer expects that he will earn $12,000 for that piece. Now the dealerís decision tree is:
So in this situation the dealer will buy the vase if
$200,000 - $8,000 Ė p > $12,000
p < $200,000 - $8,000 - $12,000
p < $180,000
The dealerís willingness-to-pay has fallen to $180,000 because of the alternative profitable opportunity that would have to be sacrificed if the vase were purchased.
Click here to find out if willingness-to-pay helps us understand hotel minibar pricing.
In summary, willingness-to-pay is a concept that relates to the situation a customer is facing. It not only depends on the direct benefits from a product, such as consumption utility, but it also depends on the alternatives that face the customer. These include the prices of related products and the customer-specific costs in consuming those products. As we will see in later segments, such interactions between different business products in a customerís decision problem play a critical role in determining the intensity of competition among those businesses.
Click here for a discussion point.
Similar to willingness-to-pay, determining willingness-to-sell involves considering the decision faced by an agent; in this case, suppliers. When suppliers choose to supply resources or inputs to a business, they are unable to supply those resources elsewhere. Therefore, suppliers essentially give up the returns they might have earned had they not supplied the business. This lost opportunity for alternative earnings is the opportunity cost incurred by suppliers. A supplierís willingness-to-sell is the minimum price they would accept and still choose to supply the resource. It is equal to their opportunity cost of supplying the resource.
To see how willingness-to-sell is determined, suppose you own and operate an ice-cream stand on weekends. You need to determine on a month-by-month basis whether or not you should continue to keep the stand open. In a typical month, you can sell 1,000 ice-cream cones on weekends. The cost of ice-cream materials is $0.20 per cone, and imagine the stand involves no cost to you. Thus, if the price per cone is p, you expect to earn (p - 0.20) x 1,000 per month. Does this mean you should stay open so long as p exceeds 0.20? If you do, you will earn some money.
The following decision tree represents your decision:
Notice that the decision tree highlights a missing variable: what does the vendor do if the stand shuts down. The vendor could work for someone else; perhaps another stand. Suppose that this employment would give $500 in wages. In this case, the vendor would be better off keeping the stand open only if (p - 0.20) x 1,000 > 500 or p > 0.70. Thus, the vendorís willingness-to-sell is $0.70 per ice cream. It is driven by the vendorís explicit costs of materials as well as their implicit cost of labour.
Click here for guided practice on how to determine economic costs.
In this topic, you have learnt
∑ how to formulate decision problems as a tree
∑ how to avoid potential pitfalls associated with sunk costs, irrelevant decisions, use of averages over margins, and accounting as opposed to economic cost
∑ how to identify a customerís willingness-to-pay
∑ how to identify a supplierís willingness-to-sell
Now go on to topic 2.3, ďDeterminants of Value in TradeĒ.