final paper man econ

11

Nonprice Competition

Learning Objectives

After reading this chapter, you should be able to:

• Identify how nonprice strategic variables can be utilized to complement pricing strategy.

• Describe how nonprice competition is a more subtle form of competition that rewards better ideas and management creativity, whereas price is a blunt instrument.

• Discuss how the Internet increases competition among firms and rewards firms who do nonprice competition well.

• Explain how nonprice strategic variables can be adjusted in theory to optimal levels.

• Describe the prisoner’s dilemma problem facing oligopolists operating under conditions of mutual dependence recognized.

©Paul Hardy/Corbis

dou70192_11_c11_325-355.indd 325 11/1/12 5:02 PM

CHAPTER 11Introduction

Introduction

In the preceding four chapters, we have been concerned with how the business firm engages in price competition, that is, how the firm should choose its price such that it expects to maximize profit (in the short run with full information) or to maximize the expected present value of profit (when revenues and costs occur beyond the present period and are not known with certainty). We saw in Chapter 4 that the firm’s quantity demanded at any price also depends on a range of other controllable and uncontrollable variables. The other variables that are controllable by the firm are the firm’s product qual- ity, its promotion expenditures, and its place of sale—these are the other three Ps of the firm’s four Ps (price being the fourth)—these are the strategic variables the firm can use to influence the demand for its product or service. In Chapter 4, we saw that changes in price will cause a movement along the demand curve, whereas changes in the other three controllable variables will cause a shift of the demand curve.

We also know from Chapter 4 that the firm’s demand depends on the concurrent actions of some other firms. Specifically, the four Ps of firms that produce substitutes (i.e., rival firms) and the four Ps of firms that pro- duce complementary goods will impact upon the focal firm’s demand. In addition, uncon- trollable customer variables such as customers’ incomes, tastes, and expectations will impact the focal firm’s demand curve when they change. All of these determinants of demand, other than the firm’s own price, are influences on demand that we mentally include when we say “other things being equal” in relation to demand curves. They are all demand shifters—when they change they cause the demand curve to shift inwards or outwards at all price levels. In this chapter, we shall consider the nonprice strategic variables (i.e., the other three Ps) that the firm can use to shift its demand curve out to the right. Thus, nonprice competition is the use by the firm of these nonprice variables, namely product design, promotion, and place of sale, to gain greater sales at any given price level.

Nonprice competition effectively focuses on differentiating the firm’s product, allow- ing the firm to increase product quality, inform and persuade customers of product dif- ferences, and offer different places of sale. In Chapter 9, we saw that the firm’s value proposition is judged by the customer as the ratio of quality over price where quality is multidimensional and is broadly defined to include all aspects of the product that the cus- tomer finds desirable, such as how it looks, how it works, what it can do, and what it does

©ASSOCIATED PRESS/AP Images

Aquafina practices nonprice competition by using nonprice variables such as product design and promotion to increase sales.

dou70192_11_c11_325-355.indd 326 11/1/12 5:02 PM

CHAPTER 11Introduction

not do (e.g., endanger or annoy). We will now argue that the other three Ps each contribute to the customer’s perception of product quality, such that nonprice competition is primar- ily about competing on the basis of quality, when quality is broadly defined to include benefits provided by promotional efforts and distribution systems.

Starting with product design, we note that it includes the shape and appearance of the product that can be expected to generate psychic satisfaction (to the extent that it is attrac- tive) to the potential customer. Product design also contributes to the product’s durability, longevity, functionality, and other aspects that the customer will perceive as utility gen- erating. For example, coffee grounds sold in an air-tight, but easy-to-open canister will be perceived as adding additional value for those who want their coffee grounds to stay fresh and yet be easily accessible. Offsetting part of (or all of) the utility from these desir- able design features might be some negative aspects of the design that give the customer disutility, such as annoying or dangerous features. For example, if the easy-to-open latch tends to break your finger nails, customers will not prefer that product and will purchase a different coffee product instead.

Product promotion typically highlights the product’s good features, informs customers of the product’s advantages (relative to rival firm’s products), and effectively congratu- lates the buyers for their good judgment in buying the product. Accordingly, promotion for a product might be expected to generate utility for the user, especially if it promotes a particular lifestyle or positive emotion or associates consumption of the product with a celebrity endorser. Thus we see, for example, billboards and TV advertisements for Coca- Cola where happy people are having a good time while drinking Coke.

Place of sale, or the distribution system utilized by the firm, also generates utility for the buyer by providing convenience both at the time of initial purchase and later when the product requires scheduled maintenance or repairs. It is much more convenient, for exam- ple, to have easy access to a local automobile dealership, rather than needing to spend a lot of time getting to and from a more-distant dealership. Online availability of many goods (e.g., hardware) and services (e.g., Netflix) with subsequent delivery by mail or by electronic downloading has made shopping much more convenient for most people. So, while we will look at promotion and place of sales separately from product design, it will facilitate discussion at times if we conceptualize more generally that nonprice competition is largely concerned with the quality element of the value proposition.1

1. This is not strictly true, of course. New places of sale will be more convenient for some buyers, where convenience of purchase can be viewed as a quality attribute, but new distribution points also will reduce the customer’s transaction costs of buying the product, which we have consid- ered as an element of the total price paid by customers. Similarly, advertising and promotion might provide information and persuasion that removes the customer’s need to conduct search costs, which we have also considered to be a component of the total price to the customer. Thus nonprice competition impacts the value proposition perceived by the prospective customer via either or both the quality or the price perception.

dou70192_11_c11_325-355.indd 327 11/1/12 5:02 PM

CHAPTER 11Introduction

Search, Experience, and Credence Goods Some products are more suited to nonprice competition than they are to price competi- tion, and vice versa. Search goods, which are defined as products for which the search cost of ascertaining product quality is relatively low, are more likely candidates for price competition, since the customers can more easily evaluate product quality. For example, your clothes are search goods; it takes only a few seconds to decide whether you like the quality (i.e., the cut and color of the fabric) of a shirt or jacket. Thus, when there is a price reduction (e.g., for Brand X shirts) the increase in the value proposition is easier for the consumer to judge. Importantly, search goods are also easier for rivals to imitate, since they can more easily see what it is that makes the product different. Thus, search goods tend to evolve towards sameness in the market as competing firms modify their product offerings to more closely resemble the products of rivals that are more profitable—this is what we saw happens in monopolistic competition (in Chapter 8) and as discussed when we considered product proliferation (in Chapter 9). As the quality of competing brands tends to gravitate toward the same level, the value proposition will be driven almost totally by the price level. Thus, firms selling search goods and seeking higher profit will tend to engage in price competition to increase their profits via larger volumes at lower levels of contribution per unit of output. Changes in design are necessary to differentiate their product again and thereby allow higher profit margins (at least temporarily until rivals also revise the design of their competing products).

Experience goods, on the other hand, are products for which product quality is more dif- ficult to judge in advance of actual consumption. Examples of experience goods are food- stuffs, restaurant meals, vacation packages, and college degrees. The difficulty to judge product quality before the purchase decision can be measured by the extent of search costs necessary to ascertain and evaluate the qualitative aspects of the product. If these search costs would exceed the price of the product, then the cheapest alternative might be for the consumer to simply buy the product and subsequently experience the quality (as we do with a new brand of cheese, for example). Alternatively, firms anxious to sell their experi- ence goods are likely to offer “taste testing” or other free or low-cost trials of the product that facilitate the consumer gaining information about product quality, and, hopefully, informing the consumer about the value proposition represented by the firm’s product. These actions by the firm are promotional strategies that serve to increase demand for the product. Note that with experience goods (that the consumer has been unable to try pre- viously) there will be uncertainty in the consumer’s mind about the product quality, and, thus, the evaluation of the value proposition will be fuzzy or indistinct, unlike the case for search goods where both the price and quality offer can be seen distinctly. Thus, if the firm cuts its price it might find the demand response is quite inelastic for experience goods (as compared to more elastic for search goods).2 Thus, firms with experience goods will tend to favor nonprice competition over price competition.

2. Recall that inelastic means that the percentage change in quantity demanded will be less than the percentage change in price, so that total revenue will increase for a price increase, for example. Conversely, elastic means that the percentage change in quantity demanded will be greater than the percentage change in price, so that total revenue will decrease for a price increase.

dou70192_11_c11_325-355.indd 328 11/1/12 5:02 PM

CHAPTER 11Introduction

Credence goods are experience goods that are likely to be different in quality the next time the customer purchases them, such as a repeat visit to a restaurant (when the chef has changed), or to an open-air concert (when the weather is inclement). Credence goods are most suited to nonprice competition, other things being equal, because the prospective buyer can less clearly see whether quality claims are indeed true until after purchase, and any prior purchases may give misleading information about product quality the next time around. For example, one’s enjoyment of a dinner while on vacation might be largely due to the attentive and knowledgeable service of a particular waiter—returning to the same restaurant the next night may be disappointing due to that waiter having the night off. Thus, the firm’s attempts to induce purchase of a credence good by reducing the price is likely to be less effective (as compared with nonprice strategies), particularly for promo- tion but also for place of sale. Promotion can be used to persuade the prospective customer that the product is of sufficiently high quality to make it the superior value proposition, while place of sales can be adjusted to increase purchasing convenience (and/or reduce transactions costs) for the customer, which can make it the best value proposition for the customer.

Nonprice Competition in the Different Market Forms In Chapter 7, we introduced the four main “market structures” which were pure com- petition, monopolistic competition, oligopoly, and monopoly. You will recall that in pure competition the products of rival firms are identical and, thus, there is no benefit for any firm to compete on the basis of product quality. But, for the other three market forms there is product differentiation across the firms. There is a relatively small degree of product differentiation for monopolistic competitors. Oligopolists typically have a greater degree of product differentiation, perhaps due to location and branding even when their core products are otherwise quite similar. Monopolists have an extremely high degree of prod- uct differentiation, since they are alone in their marketplace with no direct rivals. In each of the latter three market situations, the firm has a profit incentive to adjust its nonprice strategic variables to increase its profitability. There is an important difference, however. In monopolistic competition, the many small firms can adjust their strategic variables without expecting any direct reaction from rival firms, but in oligopoly markets the firms are few and large relative to the market and, as a result, they must recognize their mutual dependence. The gains from one oligopoly firm’s strategic actions will have direct nega- tive impact on the sales of rival firms. We should expect these rival firms to want to react with their own adjustment of strategic variables in order to win back their lost sales. The monopolist, while facing no direct competitor, may nonetheless gain from using its non- price variables to push its demand curve outwards; it might attract sales away from other products that are indirectly competitive, such as an electricity monopoly attracting sales from a gas monopoly due to one of these monopolies advertising that cooking with their source of energy is somehow better.

The More Subtle Nature of Nonprice Competition Note that price competition and nonprice competition are very different in the timing and severity of their impact on customers and on rival firms. Price competition is sudden and often quite severe in its impact, as customers will quickly switch towards (or away from) the focal firm based on their perception of the relative value propositions offered by competing firms. Price is a relatively unambiguous number, assuming that transaction

dou70192_11_c11_325-355.indd 329 11/1/12 5:02 PM

CHAPTER 11Introduction

costs, search costs, and other costs that make up the total price to the customer do not also change significantly when the firm changes its sticker price. Thus, the quality-to-price ratio calculation can be made quite quickly once the new price is known, based on the previous perception of qual- ity. The consumer will change suppliers if a new best-available value proposition becomes appar- ent, and the impact on the focal firm and on rivals will be relatively sudden.

Nonprice competition, on the other hand, must be planned before it is implemented, and its implementation will typically take much longer than it takes to implement a price change. Adver- tising and promotional campaigns must be dis- cussed, designed, and media time and space must be booked before any campaign can be implemented. Changes in product quality must be proposed, considered, and tested against the preferences of potential customers before the production facilities are modified to produce the changed product, which must then be dis- tributed to retailers or otherwise made avail- able to prospective purchasers. Similarly, place of sale, or the firm’s distribution system, cannot be changed overnight—arrangements have to be made, contracts have to be signed, facilities have to be set up, and so on.

Thus, while price competition can be decided and implemented in hours or, at most, in a few days, nonprice competition may take weeks or months to set up and implement. After that, the impact on consumers will be slower than a price change, because changes in the quality component of the value proposition will be harder to see and evaluate compared to changes in the price component. This is particularly so for experience and credence goods, of course, but even for search goods the prospective customer will have to exam- ine the quality attributes of the new product, interpret the nuances of the promotional message, or visit the new place of sale before concluding that the new value proposition offered is superior to rival offerings, or not.

Nonprice competition is therefore typically less abrupt and more gradual in its impact (on the firm’s sales) than is price competition. It needs to be considered and planned well in advance of its implementation, such that quick nonprice retaliation in response to a rival’s nonprice initiative is usually impossible. This means that a well-planned and well- implemented nonprice strategy can gain a market advantage that endures for quite a while before rivals are able to mount their own new promotional campaigns, product design changes, or new distribution outlets. During that time, the focal firm will enjoy increased sales at the same (or higher) price levels until rival firms either reduce their prices or come up with their own nonprice strategic initiatives. Indeed, during the time that rivals are

©Digital Vision/Thinkstock

Price competition is sudden and requires little advanced planning whereas nonprice competition takes time to plan before it can be implemented.

dou70192_11_c11_325-355.indd 330 11/1/12 5:02 PM

CHAPTER 11Section 11.1 Demand Shifting Using Nonprice Strategic Variables

trying to catch up with the focal firm’s nonprice strategic initiative, that firm can be work- ing on its next nonprice initiative to once again shift its demand curve outward.

In the remainder of this chapter, we will examine changes in product design, promotion efforts, and place of sale (distribution systems) to see how the firm can increase its profit- ability by changes to these nonprice strategic variables.

11.1 Demand Shifting Using Nonprice Strategic Variables

As noted, the aim of nonprice competition is to shift the firm’s demand curve to the right, to induce more sales at the current (or any other) price level. This is not a limitless process, however. After some point, additional product design features, promotional expenditures, and distribution outlets will cost more to set up and operate than they will generate in terms of additional sales. The manager’s problem is to increase the expenditures associated with these nonprice variables to an appropriate level such that the firm’s profit (or expected net present value) is maximized. In this section, after a short review of the nonprice variables, we shall consider how the firm might adjust these to maximize its profit in the short run (under full information) or maximize its expected net present worth (under uncertainty and longer time horizons).

Product Quality The multiple dimensions of product quality include how the product looks (e.g., its shape, appearance, or design aesthetics); what it does (e.g., its functionality, or usefulness); how long it lasts (e.g., its durability, or robustness); and what it does not do (e.g., it does not endanger, annoy, or cost more money to maintain). Improving product quality means increasing the first three of these aspects of quality and reducing the latter. In Chapter 3, we considered product attribute analysis, which identified the main product attributes (or quality dimensions) that consumers will use to differentiate between and among com- peting products and on which they base their purchase decision. Put another way, the product attributes of importance to each consumer are the ones that enter the quality component of the firm’s value proposition that is perceived by that consumer.

Thus, the firm’s managers need to view the product through the eyes of their customers to appreciate which of these quality dimensions are more (or less) appreciated by those customers and by prospective customers. Market research, which is asking customers and prospective customers what they like and do not like about the product, will serve to inform management as to which of the qualitative features are particularly desirable and which might be increased, reduced, added, or deleted (Kim & Mauborgne, 1999).

Packaging and People

As first mentioned in Chapter 3, some marketers talk about the six Ps of marketing, add- ing packaging and people to the traditional four Ps. Packaging refers to the way in which a product or service is presented to the potential customer. For example, attractiveness and the security of the packaging can be viewed as aspects of product quality. The pictures on the boxes in which new TVs are delivered raise the customer’s anticipation of the viewing

dou70192_11_c11_325-355.indd 331 11/1/12 5:02 PM

CHAPTER 11Section 11.1 Demand Shifting Using Nonprice Strategic Variables

experience and the strength and rigidity of the box and interior packaging ensures that the TV will work immediately when it is turned on. The term people relates to the human element, or the quality of personal service associated with the purchase and delivery of the product. Note that by “product” we mean “product or service” and that both physical products and intangible services can be delivered with better (or worse) personal service by the provider. This personal service is often inextricably combined with the product, of course, and even when it is a minor component of perceived product quality (such as for impulse purchases of commodity items) it is rarely irrelevant. Thus, what we have said above about adjusting quality dimensions to the optimal levels also relates to packaging and personal service—these should be augmented to the extent that the marginal cost of additional packaging and service is just equal to the marginal revenue deriving from that additional packaging and service. For example, a restaurant manager considering whether to hire an additional waiter (for say, $100 per day) must consider whether the improved service provided will generate at least $100 additional contribution to over- heads and profit from additional food and beverages ordered either immediately or via repeat purchases of satisfied customers.

Promotion Promotion includes a variety of activities that are designed to induce the prospective cus- tomer to purchase the firm’s product or service. Advertising is perhaps the most com- monly used promotional tool, but also Internet websites, Google advertising and key-word purchase, point-of-sale displays and give-aways, and support of sporting events are also common promotional vehicles.

Many firms have an advertising budget, often calibrated to be a particular proportion of sales revenue, to be spent on advertising or other promotional activities. For example, Coca-Cola is reported to spend about 30% of its revenue (or wholesale price per can) on advertising in order to maintain its market share in the face of advertising by rival firms.3 Managers will be concerned whether or not the current level of promotional expenditure is the profit-maximizing level. They should expect that increased advertising expenditures will lead to increased sales of their products, assuming, of course, competent advertis- ing campaigns that do not “turn off” customers. But they should also expect diminish- ing returns to advertising expenditures, such that there is an optimal level of advertising expenditures for their particular product and market situation. Assuming that Coca-Cola has gravitated over time to an optimal level of advertising expenditure, any reduction in their advertising is likely to be accompanied by a reduction in both sales and profits.

In oligopoly markets, the firm will be concerned about the level of its advertising expendi- tures relative to the advertising expenditures of rival firms. If rivals increase their advertis- ing the focal firm should expect some erosion of its own market share as rivals’ demand curves shift outwards and its own demand curve shifts inwards. A large fraction of the oligopolist’s advertising budget may be required simply to maintain its market share.

3. Firms need to keep their brand name product quality assertions in the minds of consumers, not only to prevent direct rivals (e.g., Pepsi and RC-Cola in Coca-Cola’s case) but to avoid being replaced by indirect rivals (other beverages) and to avoid being forgotten amongst the “noise” of many other firms advertising their unrelated products.

dou70192_11_c11_325-355.indd 332 11/1/12 5:02 PM

CHAPTER 11Section 11.1 Demand Shifting Using Nonprice Strategic Variables

Thus oligopolists keep a sharp eye on the advertising campaigns of rivals and will feel compelled to increase their own advertising (or adjust another strategic variable) to coun- ter increased advertising by rivals. In most oligopoly markets, the firms will gravitate towards a level of advertising that seems to provide the optimal contribution to overheads and profit, and then watch their rivals to see that they maintain the status quo.

A complicating issue for the evaluation of advertising expenditures is that advertising campaigns are likely to differ qualitatively and in their impact on consumers—some advertising concepts will be more successful than others, and some campaigns will be interrupted by adverse weather or major events that distract consumers and reduce the effectiveness of the advertising campaigns. For example, during the Olympic Games, or the presidential election campaign, advertising by the firm might be overlooked by cus- tomers more interested in the sporting or political events. Similarly, new health informa- tion, such as new data relating obesity to the sugar and fat content of fast food, may render advertising less effective in increasing the firm’s demand. Where advertising can be increased by more of the same promotional message (e.g., more billboards in new loca- tions utilizing the same promotional message, or the same advertisement placed in more newspapers) the firm’s managers can gain a more reliable estimation of the marginal impact of increased advertising on the demand for their products.

Internet Promotion

Quickly replacing billboards, print media, radio and tele- vision advertisements, and even point-of-sale promotional materials as the media of choice for many firms is Internet promotion, informative and persuasive material posted on the firm’s (and other) websites, cross-posting with other firm’s websites, purchase of Google “ad-words” and paid advertise- ments that pop-up when people search for specific information on the Internet. Social media sites, such as Facebook, allow the firm to have ongoing com- munication with customers and potential customers, and also allow the firm’s products, service quality, and corporate social responsibility to be rated by millions of customers and, thus, provide information to potential customers about the value proposition offered by the firm.

Increasingly, people search for information just-in-time on the Internet, that is, just before they plan to use the information, rather than keep printed advertisements and other mate- rials on hand for reference just-in-case they will need this information at a later time. It stands to reason that the immediate availability of product price and quality information

©ASSOCIATED PRESS/AP images

Internet promotion like Google advertising is designed to induce the prospective customer to purchase a firm’s product or service.

dou70192_11_c11_325-355.indd 333 11/1/12 5:03 PM

CHAPTER 11Section 11.1 Demand Shifting Using Nonprice Strategic Variables

via Internet websites tends to reduce the impact of other forms of just-in-case promo- tion. Whereas print media, radio, and television provide seller-supplied (and potentially biased) information about the product, the Internet allows the prospective customer to quickly find (and compare) offers from various sellers and usually also to find reviews and ratings of the firm’s product in comparison with rival products. These reviews and comparative ratings are particularly effective for the promotion of experience and cre- dence goods, assuming the firm’s product is rated highly, of course. Thus, the Internet has not only impacted the efficacy of other forms of promotional media but has also served to increase the quality of many firm’s products since any adverse comparisons or deficien- cies are quickly noted and publicized by consumers and other interested third parties. Indeed many companies explicitly ask (on their websites) for ratings and reviews of their products to provide them with the information they need to improve product and service quality.

Place of Sale The place of sale refers more broadly to the firm’s distribution system, which includes the way in which the product is made available for viewing or consideration by prospective customers, and the means by which it is delivered to the customer at or after the point of sale. Thus, the firm might set up physical shops where the product can be viewed, pur- chased, or consumed by customers who travel to the shop and purchase the product there. Heavy or bulky items may need to be delivered subsequently to the customer’s place of residence or business, or the purchaser may take delivery at the point of purchase, con- sume it there (consider the case of services, food, and beverages, for example), or trans- port it themselves to the desired location.

As mentioned elsewhere, the place of sale creates more or less inconvenience, search, and transaction costs for the prospective buyer. This will enter the prospective customer’s evaluation of the value proposition offered by the firm, either as a transaction or search costs on the (price) denominator, or as a product attribute on the (quality) numerator of the value proposition. By having multiple places of sale, the firm reduces transactions and search costs and increases the convenience for at least some customers and, as a result, may become the best value proposition for some of those customers, thus selling more units of output.

In Chapter 9, we considered franchising as a means of gaining additional market share and greater firm-level profit even under conditions of monopolistic competition where the individual franchise might make only normal profits. By opening additional fran- chises, the firm is effectively making its place of sale more convenient while reducing the search and transaction costs for more people and, thus, gaining customers that previously did not perceive that firm to be offering the best value proposition.

Internet Sales

As implied earlier, the physical store is declining as the major place of sale for many goods and services, being progressively replaced by the Internet websites of firms where prospective customers can view images of the product, gain information about product attributes and comparisons with rival products, and then purchase and pay online, with delivery subsequently to the customer’s desired location within hours, in some cases,

dou70192_11_c11_325-355.indd 334 11/1/12 5:03 PM

CHAPTER 11Section 11.2 The Optimal Level of Nonprice Competition

and, generally, within a few days. Prospective customers, increasingly “time-poor” and faced by an increasing array of rival products (including imports or products that would be imported directly to the consumer after purchase), find that their transaction costs are reduced significantly by Internet purchases. Thus, buying online offers a better value proposition even when buying the same item from the same seller at that firm’s physical store. In addition to the savings of transaction costs, the customer may find that Internet prices are lower than in-store prices for two main reasons. First, the costs of displaying and selling the items in-store are likely to be higher than selling via the Internet and ship- ping from a warehouse. Second, the customer who is in the store has already invested time (and possibly also parking costs) and would need to repeat this investment in other stores to find comparative price and quality information. That person is likely to make the economically rational decision to buy in store at what is likely to be a higher price than may be available in other stores or on the Internet rather than incur additional search costs to find out if that is in fact true.

11.2 The Optimal Level of Nonprice Competition

It is useful to consider the adjustment of a nonprice strategic variable in two stages. First, for simplicity of exposition, we shall consider adjusting the nonprice vari-able while holding price constant, which is appropriate for some market situations. Later we shall adjust price and the nonprice variable simultaneously to find the profit- maximizing combination of price and nonprice strategies, which is appropriate for other market situations. We saw in Chapter 7 that firms operating in oligopoly markets with mutual dependence recognized often face price rigidity—they will be reluctant to raise their price for fear of the elastic demand response that would happen if their rivals did not also raise their prices, which would cause their market share and profitability to decline substantially. They might also be reluctant to reduce their price for fear of an inelastic demand response, which would happen if all rivals also reduce their price to protect their market share, such that while their market share would stay the same, each firm’s profit margin on each unit sold would be substantially less. In such kinked-demand-curve mar- kets the oligopoly firm effectively faces a price that it prefers to keep fixed at the current level.4

We might also expect price to be held constant in monopoly markets where the firm’s price is regulated by a government authority and is, thus, set at a particular level that is main- tained for a considerable period between regular (perhaps annual) reviews of the price level by the regulatory authority. As indicated earlier, public utilities, such as the post office, electric company, and gas suppliers, are often regulated in the public interest since

4. As we saw in Chapter 6, the short-run profit-maximizing firm will want to change its price if its MC curve intersects a concrete section of the disjointed MR curve that accompanies the kinked demand curve, and will tolerate the market share loss because its objective is to maximize prof- its in the short run, and thus the subsequent-period benefits of market share (such as repeat sales) are not included in the firm’s objective function. When oligopolists consider the longer run impacts of their pricing decisions they are likely to hold price constant to maintain market share for future period benefits, assuming they are not practicing price leadership or followship, of course.

dou70192_11_c11_325-355.indd 335 11/1/12 5:03 PM

CHAPTER 11Section 11.2 The Optimal Level of Nonprice Competition

there are no competing firms to induce the firm to be more effi- cient in production and more competitive with their prices. Without competition, monopo- lies tend to allow their cost struc- tures to inflate such that prices would periodically increase, so public regulation of prices is uti- lized to prevent upward creep in their prices due to managers’ unwillingness to pursue pro- duction efficiencies.

So, in the following section, we shall first consider the most sim- ple case where the firm leaves price at its current level and adjusts one of the other strategic

variables to the profit-maximizing level. Although this could be any one of the three non- price strategic variables, in the following we will discuss the process in terms of product quality, and later generalize our findings to refer to all other nonprice strategic variables. If the firm were to increase its product quality by constant increments to expenditure (e.g., $1,000) we should expect the demand curve to shift outward but we should also expect to observe diminishing returns to market demand for the product quality. That is, for a series of increases to product quality (with each increase costing $1,000), we should expect the outward shift of the demand curve to be progressively less, as shown in Figure 11.1.

Figure 11.1: Diminishing returns to quality

$/Q

Price

Quantity/period (Qt)

Various demand curves, shifting outwards due to increasing increments to product quality, with each increment costing e.g. $1,000.

©Images.com/Corbis

Utility companies such as electricity and gas suppliers, are often regulated in the public interest, since no competing firms exist to stimulate increased efficiency in production and more competitive pricing.

dou70192_11_c11_325-355.indd 336 11/1/12 5:03 PM

CHAPTER 11Section 11.2 The Optimal Level of Nonprice Competition

Figure 11.1 demonstrates diminishing returns to product quality—the demand curve shifts outwards by progressively less for each constant increment to expenditures on product quality. Because total revenue (TR) is equal to price times quantity (i.e., PQ) and since price (P) is constant, it is evident that TR must be increasing at a decreasing rate because the quantity (Q) is increasing at a decreasing rate (as advertising is augmented by constant increments). This means that the marginal revenue associated with quality augmentation must be falling. But, in addition, we should also expect to find the marginal cost of qual- ity augmentation is increasing, due to diminishing returns to expenditures on product quality. That is, constant increments (e.g., $1,000 increases) to expenditures on quality will increase the level of quality by diminishing amounts. This means that the marginal cost of quality augmentation (i.e., the cost per unit of additional quality) will be increasing.

In Figure 11.2, we show the profit-maximizing level of quality augmentation—it will occur when the (rising) marginal cost of quality augmentation just equals the (falling) marginal revenue due to quality augmentation. The marginal cost and marginal revenues associ- ated with quality augmentation are shown in Figure 11.2 as the slopes of the total curves, respectively. The curve TCqa (for total cost of quality augmentation) rises at an increasing rate as quality is increased, while the curve TRqa (for total revenue from quality augmen- tation) rises at a decreasing rate as quality is increased. These are the total cost and total revenue associated with changing quality above the initial level (and do not include the production cost of the basic product). The slopes of these curves represent the marginal cost and the marginal revenue associated with augmenting advertising. These slopes are equal at the quality level shown as K*, which is the profit-maximizing level of quality, since the marginal cost of increasing quality is just equal to the marginal revenue due to the increase in quality, all other things being held constant.

This result is the (by-now-familiar) marginal-equality condition of economics—for profit maximization any strategic variable should be adjusted to the point that the marginal revenue obtained from that adjustment is just equal to the marginal cost associated with that adjustment. In this case it is profit-maximizing to raise quality to the level where the marginal cost of quality augmentation is just equal to the marginal revenue due to quality augmentation. At quality levels below K* the marginal revenue exceeds the marginal cost so it increases profit to augment quality up to this level. At quality levels above K* the marginal cost exceeds the marginal revenue from quality augmentation so profit would fall if quality was augmented to these levels. Thus, the profit-maximizing level of quality is K* where the last dollar spent on quality augmentation is just covered by a dollar con- tributed by the additional sales of the product in the market.

dou70192_11_c11_325-355.indd 337 11/1/12 5:03 PM

CHAPTER 11Section 11.2 The Optimal Level of Nonprice Competition

Figure 11.2: The quality augmentation decision

Total Cost & Revenue due

to quality augmentation

($)

Marginal Cost & Marginal Revenue of

quality augmentation

($)

Slope = MRqa

Slope = MCqa Quality level

Quality level

TRqa

TCqa

MCqa

K*

K*

MRqa

It is important to note that the marginal cost of quality augmentation as shown in Figure 11.2 will be composed of two separate cost categories that are likely to shift upwards as the level of quality increases. The first is an increase in fixed costs. Additional machines and specialized equipment might be required to allow the firm to produce higher quality products, so we might imagine the total fixed cost (TFC) curve shifting upwards continu- ally (or jumping up at points where better equipment is needed to improve quality fur- ther). Secondly, the total variable cost (TVC) curve is likely to shift upwards as quality is continually increased, because more labor time and higher quality raw materials or com- ponents are required to increase quality progressively. Thus the MCqa curve in Figure 11.2 is the change in both TFC and TVC that is required to raise quality by an additional unit (of quality). You can see that the analytical process underlying Figure 11.2 is a notional planning exercise conducted to find the optimal level of quality. Once that level has been found (in theory) or estimated (in practice) the firm can select the appropriate combina- tion of machines and equipment that can provide the desired level of quality and, thus, have a particular level of TFC. Similarly, it will utilize the appropriate level of labor per unit of output and quality of raw materials and components such that the desired level of quality is attained. The firm would have a particular TVC curve that it would move along as quantity of production changes while quality is constant at the chosen level.

We know that product quality is multidimensional, so we must note that the above con- ceptual analysis applies to only one particular dimension of product quality. For example, the dimension under consideration could be package size. Suppose a new firm that is

dou70192_11_c11_325-355.indd 338 11/1/12 5:03 PM

CHAPTER 11Section 11.2 The Optimal Level of Nonprice Competition

planning to produce fresh orange juice can only afford one bottling set-up and wants to choose the best size of a plastic container for its orange juice. It will consider packaging its juice in, for example, 1-quart, 2-quart, 3-quart, or 4-quart bottles, or indeed any fractional volume along the spectrum from 1 quart to 4 quarts. The above analysis would allow the optimal container size to be determined. In practice, of course, the firm needs to consider the market reality that orange juice is normally presented to consumers in standard pack- aging sizes (e.g., 1- and 2-quart containers), so it may wish to choose whichever package size is superior in terms of profit generation. Alternatively, it may decide to offer a dis- tinctly different container size (e.g., 1.35 quarts) and in its promotion claim that this size provides “extra value” for the customer if it believes that this is the profit-maximizing combination of nonprice strategies.

In theory, the firm would repeat this analysis for each dimension of product quality that is recog- nized by its customers. For example, the sweet- ness of the orange juice can be adjusted by the choice of orange variety as a raw material or by the addition of sugar or other sweetener. Simi- larly, the color could be adjusted by the addition of minute quantities of food dye, or the fraction of solid content (pulp) in the juice can be adjusted by calibrating the juicing process of raw oranges. Each one of these quality dimensions for orange juice must be considered and set by the firm at the optimal level, with a view to maximizing its short-term profit or net present value of profit.

In practice, the firm’s managers would adjust each dimension to the levels indicated by their market research. Conjoint analysis is a market research technique that allows customers to conjointly value (i.e., consider together, relative to each other) the quality dimensions of any product and thus, managers can determine how much each quality attribute should be augmented or dimin- ished (Green & Srinivasan, 1978). Just as the reser- vation price of customers varies, each respondent to a marketing survey will potentially value the quality dimensions (which we called product attri- butes in Chapter 3) of the product somewhat dif- ferently. These customer valuations will reflect how much extra the customers would, on average, pay for an additional unit of each quality dimension. The manager will consider the average customer valuation and compare this with the cost of augmenting quality and, subsequently, choose the level of quality for each particular product attribute. For example, Kia Motor company introduced a six-speed transmission and a smaller turbo- diesel engine to its range of automobiles to allow enhanced acceleration and increased fuel economy, after concluding that its customers wanted better acceleration and better economy.

©Martin Poole/Thinkstock

From taste to color, each quality dimension of orange juice must be considered and set by the firm at what it thinks is the optimal level to maximize profit.

dou70192_11_c11_325-355.indd 339 11/1/12 5:03 PM

CHAPTER 11Section 11.2 The Optimal Level of Nonprice Competition

Simultaneous Adjustment of Price and Quality For the monopolistic competitors (and for the monopolist whose price is not regulated), the fact that they can set prices without fear of rival reactions means that the increase in demand due to quality augmentation also means that they can raise their price to find the profit-maximizing price and quality level simultaneously. To show this graphically we need to develop several new concepts and then bring them together in one (admittedly more complex) diagram. But you will be ready for this level of complexity because each of the concepts is relatively simple and the analysis utilizes knowledge that you have learned in the preceding chapters.

We start with the locus of optimal prices (LOP) which is a line joining the optimal (i.e., profit-maximizing) price level at each different quality level. In Figure 11.3 we show two demand curves (and related MR curves) that reflect two different levels of quality—D1 and MR1 reflect one quality scenario while D2 and MR2 reflect a higher level of quality scenario. We also show two marginal cost of production curves, MC1 and MC2, which relate to two different levels of quality that might be chosen. The optimal price in the first quality scenario is P1, and the optimal price in the second quality scenario is P2. (Note that the optimal output level is found where MR 5 MC in each scenario, and these prices are found by projecting up to the relevant demand curve from those output levels in each scenario.) The line joining those two optimal price levels, shown as LOP, is the locus of these profit-maximizing price and output combinations (and many other optimal price and quantity combinations). We need this locus showing all possible optimal price and quantity combinations so that we can superimpose it on the cost side of the quality aug- mentation issue, which we now address.

Figure 11.3: The locus of optimal prices

Quantity/ period

P2

MR2MR1 D1 D2

MC2

$/Q

LOP

P1

Q1 Q2

MC1

dou70192_11_c11_325-355.indd 340 11/1/12 5:03 PM

CHAPTER 11Section 11.2 The Optimal Level of Nonprice Competition

The locus of average quality costs (LAQC) is a line joining the average quality cost at each profit-maximizing price level. The average quality cost (AQC) is the total costs of qual- ity at a particular level of quality, divided by the number of units of output subsequently produced and sold. Total cost of quality can be regarded as a fixed cost once the product design (and thus the level of quality) is chosen and the firm’s plant is set up for that qual- ity level. Subsequently, when output is produced, the total costs of quality is a constant, and thus, the average quality cost curve will be a rectangular hyperbola, as shown in Fig- ure 11.4, for each of the two quality scenarios. In each scenario AQC values will be very high initially but will fall progressively as the total cost of quality is divided by a progres- sively increasing level of output. As you can see in Figure 11.4, in the first quality scenario the optimal output level was Q1 (from Figure 11.3) and the average level or quality costs per unit at that output level is A1. For the second quality scenario, the average quality cost is A2 at output level Q2. The line drawn through these AQC and Q coordinates will be a locus of average quality costs, shown as LAQC, which represents the AQC value at each of the optimal output levels associated with each quality scenario.

Figure 11.4: The locus of average quality costs

Average Quality

Cost per unit ($)

Quantity/ period

A2 A1

Q1 Q2

AQC2

LAQC

AQC1

On the basis of the LOP and LAQC curves we can now demonstrate the simultaneous determination of the profit-maximizing quality (K), output (Q) and price (P) levels. Note that each of the locus curves in Figures 11.3 and 11.4 represents an average, and there must be a marginal curve associated with each of these average curves. These marginal curves would show the rate at which the total in each case was changing—that is, the locus of marginal quality costs (LMQC) will show the rate at which total quality costs are changing as quality changes (see Figure 11.2), and the locus of optimal marginal revenue (LOMR) will show the rate at which the total revenue is changing as quality changes (see Figure 11.2). Whereas Figure 11.2 was in two dimensions (i.e., quality vs. cost/revenue) holding output level constant, we now need to find the optimal output level while allowing both quality and the cost/revenue dimensions to vary. Rather than show a three-dimensional

dou70192_11_c11_325-355.indd 341 11/1/12 5:03 PM

CHAPTER 11Section 11.2 The Optimal Level of Nonprice Competition

diagram, in Figure 11.5 we show the marginal and average locus curves for both prices and quality costs mapped against the third variable of interest, the output level. At the output level where the marginal curves intersect (i.e., LOMR 5 LMQC), we sketch in the demand curve (D*) and the AQC curve (AQC*), both of which must cross the LOP and LAQC curves, respectively, at that output level. These show the profit-maximizing price and quality levels that the firm should set to maximize its profit. The profit-maximizing price will be P*, the profit-maximizing level of average quality costs will be AQC*, and the profit-maximizing level of output will be Q*. The total expenditure on quality augmenta- tion will be the AQC* times the output level Q*.

Figure 11.5: Simultaneous choice of price and quality levels

Quantity/ period

P*

AQC*

Q*

AQC*

D*

LOP LOMR

LMQC

LAQC

$/Q

Thus we have demonstrated that the optimal levels of price and quality can be determined simultaneously to solve the manager’s problem of finding the optimal value proposition for the firm. The above model serves to demonstrate the principle that price and any one of the nonprice strategic variables can be adjusted simultaneously to increase profitability. Indeed, in theory, price and all of the other strategic variables (and all of their composite dimensions) can be adjusted simultaneously to best position the firm’s product in its mar- ket, but to prove that here would require mathematics of a fairly high order and is best left alone! In practice, of course, managers will not have the information necessary to draw in all these curves and see where they intersect, and so will need to utilize an iterative pro- cedure to approach this optimal price and quality combination by trial and error and the exercise of judgement based on their knowledge of consumer behavior in their markets.

Estimating the Optimal Level of Nonprice Strategic Variables The foregoing sections have established the theory of optimal nonprice competition and have demonstrated the principles involved in adjusting several strategic variables simul- taneously to find the profit-maximizing combination of those strategic variables. In prac- tice, the firm will find that the information search costs required to utilize these theoretical

dou70192_11_c11_325-355.indd 342 11/1/12 5:03 PM

CHAPTER 11Section 11.2 The Optimal Level of Nonprice Competition

models would be prohibitive. Nonetheless, an understanding of the theory of nonprice competition will allow the manager to use parts of the theory (for which information is available) to estimate the profit-maximizing level of particular nonprice strategic variables.

For example, a firm that consistently advertises could, at relatively small cost, collect time- series data on monthly quantity demanded (in thousands) and advertising expenditures per month (in thousands of dollars), and could calculate a line of best fit between these two sets of data observations. Suppose a firm has done this and utilizing correlation anal- ysis (see Chapter 4) has found the line of best fit to be Qd 5 a 1 bA where A represents advertising expenditures in thousands of dollars. The coefficient to advertising in this line of best fit, that is, b, is the marginal impact on quantity demanded for a $1,000 change in advertising expenditure. It is a simple matter to then calculate the marginal revenue due to a $1,000 change in advertising5 and compare it with the marginal cost (i.e., $1,000) of a unit of expenditure on advertising. The statistic b will be some fraction or multiple of a unit of output. Let us suppose it is 0.25, meaning that an additional $1,000 spent on adver- tising increases sales by one quarter of 1,000 units (i.e., 250 units) of the product. Now, if the contribution to overheads and profit of the product is say, $5, those 250 units generate $1,250 in contribution, which exceeds the $1,000 marginal increment in advertising expen- ditures. Thus, advertising could be increased to maximize profit, assuming that all other influences on quantity demanded remain the same. Perhaps several more increments of $1,000 could be spent, until the marginal revenue due to advertising falls to equality with the ($1,000) cost of the marginal unit of advertising.6

Using multiple regression analysis, the manager can control for other variables that are likely to have changed to better isolate the impact of advertising on quantity demanded, as we saw in Chapter 4. Also, multiple regression analysis allows us to estimate nonlinear relationships between quantity demanded and the nonprice variables. We have argued above that there will almost certainly be diminishing returns to these nonprice variables. As an example, suppose the demand function for a firm has been estimated as:

Q 5 10,000 1 25.2 A – 0.8 A2 (11-1)

Where Q is quantity demanded in units and A represents advertising expenditures in thousands of dollars. The optimal level of advertising will be the level at which the last $1,000 spent on advertising contributes just $1,000 towards overheads and profit. Thus the maximizing condition is d/dA 5 1 where  (the lowercase Greek letter pi) represents the contribution to overheads and profit. To find d/dA we must first find out how quantity

5. The marginal revenue will be the increase in quantity demanded multiplied by the contribution per unit, assuming that price and AVC remain constant at the higher output level.

6. Although the line of best fit is a linear expression, we expect that there will be diminishing returns to advertising expenditures, so the marginal impact on quantity demanded of $1,000 of advertising is not likely to remain at the value b 5 0.25. Thus the firm must monitor the impact of additional advertising on its quantity demanded, and stop increasing advertising when the incremental revenues just equal the incremental costs.

dou70192_11_c11_325-355.indd 343 11/1/12 5:03 PM

CHAPTER 11Section 11.2 The Optimal Level of Nonprice Competition

demanded varies with advertising (i.e., dQ/dA) and then how profit varies with quantity demanded (i.e., d/dQ, which is the contribution margin). Thus d/dA expands to:

d/dA 5 dQ/dA ? d/dQ (11-2)

From equation 11-1, we can find dQ/dA by taking the first derivative of the estimated demand function with respect to advertising expenditures, which is:

dQ/dA 5 25.2 – 1.6A (11-3)

Since the profit-maximizing condition is to set d/dA 5 1, and because d/dQ is the contribution margin (CM), we can restate equation 11-2 as:

dQ/dA 5 1/CM (11-4)

Thus, the profit-maximizing rule is to set the marginal impact of advertising on quantity demanded equal to the reciprocal of the contribution margin. From the estimated demand function (11-1), we found the marginal impact of advertising on quantity demanded as shown in equation 11-3. Now assuming that the contribution margin is $6 per unit, we can solve for A by setting equation 11-3 equal to 1/6 as follows:

25.2 – 1.6A 5 1/6

Multiplying both sides by 6 we find:

151.2 – 9.6A 5 1

Taking 151.2 from both sides we have:

– 9.6A 5 – 150.2

Finally, by dividing both sides by –9.6, we find that A 5 15.646. Thus, when the contribu- tion margin is $6 per unit of output, the estimated optimal level of advertising is $15,646.

In the above example, we implicitly assumed that the contribution margin was constant at $6, regardless of the output levels, and that the shifting outward of the demand curve due to advertising would allow all additional units of output to be sold at the same price level. This is a simple case that, nonetheless, might be applicable (or sufficiently close to reality) in many business situations. But, where there are diminishing returns in produc- tion (i.e., rising MC as output increases), the contribution margin will fall as output levels increase so this must be incorporated into our calculations. Recalling that the contribution margin is equal to price (P) minus average variable costs (AVC), and that a rising MC causes the AVC value to rise, we would have an expression for contribution margin of the form CM 5 P – AVC – (AVC/Q), where the latter term indicates the extent to which AVC is expected to rise as output increases. But you will recall from Chapter 5 that as the output rate is increased, AVC falls at first; bottoms out where MC cuts the minimum value of AVC; then rises again as the MC continues to rise. Thus, if the firm is operating at an output level anywhere near the minimal point of its AVC curve, the AVC curve will

dou70192_11_c11_325-355.indd 344 11/1/12 5:03 PM

CHAPTER 11Section 11.2 The Optimal Level of Nonprice Competition

be relatively flat on each side of that minimal point, and so the assumption of a constant contribution margin may be a sufficient approximation for calculating the optimal level of advertising.

Where managers do not have inexpensive access to sufficient data on the relationship between prior advertising (or other nonprice variable) and the quantity demanded, they might experiment by implementing a slight increase or decrease in advertising expendi- tures (or other variables) to observe by how much sales volume changes as a result of the change in that nonprice strategic variable and thus, gain an appreciation of whether they should increase or decrease that strategic variable further in pursuit of the maximization of their firm’s net present worth.

Nonprice Competition With Mutual Dependence Recognized In oligopoly markets, the firm should realize that its adjustment of strategic variables will have noticeable impact on the sales of rival firms and consequently that these other firms are likely to react by adjusting their own strategic variables and thus impacting the focal firm’s sales, trig- gering another round of adjustments, and so on. When the strategic variable being adjusted is price, this sequence of action and reaction could result in a price war, whereby the prices of all firms are adjusted downward repetitively until no firm is making any or much profit. Clearly, profit-maximizing firms will want to avoid this outcome, resulting in the emergence of price lead- ership and price fixing agreements, as mentioned in Chapter 8.

We have noted that, unlike price competition, it typically requires a significant lead time to imple- ment a change in a nonprice strategic variable. As a result, if an oligopolist is caught napping by a rival’s new promotional campaign, product improvement, or new retail outlet, there will be a significant lag before it can retaliate effectively, during which time it may have lost a significant part of its market share that may be very hard to regain. The existence of this lag, therefore, motivates firms to maintain an ongoing involvement in promotional activity—if the firm is always planning its next promotional campaign, it will not be caught napping to the extent that it would be if it had to start from scratch after a rival launches a new promotional campaign.

The threat that a rival might launch a major nonprice strategy and gain market share at the expense of the focal firm may cause the mutual-dependence-recognized oligopolist to

©ASSOCIATED PRESS/AP Images

A price war occurs when the prices of all firms are adjusted downward repetitively until no firm is making any or much profit. To avoid a price war, firms can create price leadership and price fixing agreements.

dou70192_11_c11_325-355.indd 345 11/1/12 5:03 PM

CHAPTER 11Section 11.3 Game Theory in Mutual-Dependence-Recognized Oligopoly

increase its advertising, quality, or distribution outlets as a defensive move, just in case a rival firm does the same thing. In effect the oligopolist is involved in a strategic game in which the rival’s next move is anticipated, so the firm will be induced to make an aggres- sive nonprice adjustment in order to render less potent the rival’s move if indeed it occurs.

11.3 Game Theory in Mutual-Dependence-Recognized Oligopoly

A game is a situation in which two or more players choose strategies to compete for a reward or payoff of some kind. In mutual-dependence-recognized (MDR) oligopoly the game is producing and selling a similar product, the strategy is (let’s say) advertising expenditures, and the payoff is market share and the profit associated with that market share. A game that contests market share is a zero-sum game, meaning that the gains of some players equal the losses of the other players. In Table 11.1 we show the payoff matrix for a two-person, zero-sum game in which the players are Firm A and Firm B. We assume that they are each considering two alternative strategies—strategy 1 is to maintain their advertising at $4 million per year, and strategy 2 is to increase their advertising expenditure to $6 million per year. The four quadrants of Table 11.1 show the four possible combinations of the two firms’ strategies. The numbers in the top-left-hand quadrant show the payoffs to the two firms if both maintain advertising at $4 million each—by convention we show A’s payoffs first, followed by B’s payoffs, in each quadrant. When both firms spend $4 million on advertising the expected profits are shown to be $10 million for each firm.

Table 11.1: Payoff matrix for two-person game

Firm B’s Advertising Budget

$4m $6m

Firm A’s Advertising Budget $4m $10m; $10m $5m, $12m

$6m $12m, $5m $8m, $8m

Suppose now that Firm A were to increase its advertising expenditure to $6 million while Firm B maintains its advertising at $4 million. The payoffs for this scenario are shown in the lower-left-hand quadrant: Firm A’s profits increase to $12 million while Firm B’s profits decline to $5 million. This result occurs because the additional advertising of Firm A causes some customers to be persuaded that Firm A offers the better value proposition and as a consequence they switch their purchases from Firm B to Firm A. Oppositely, sup- pose that it was Firm B that increased advertising to $6 million while Firm A maintained advertising at $4 million (the payoffs for this scenario are shown in the top-right-hand quadrant, and are $5 million to Firm A and $12 million to Firm B). Finally, suppose that both firms increase advertising to $6 million, the payoffs (shown in the lower-right-hand quadrant) are $8 million to each firm. This result occurs because the increased advertis- ing of each firm essentially offsets the impact of the other firm’s increased advertising, but while the increased advertising does attract new buyers to both firms, the shift of their demand curves does not increase total revenue by enough to cover the additional $2

dou70192_11_c11_325-355.indd 346 11/1/12 5:03 PM

CHAPTER 11Section 11.3 Game Theory in Mutual-Dependence-Recognized Oligopoly

million in advertising expense that each firm incurs. This is the worst possible outcome, since profits have fallen compared to the initial situation where $4 million advertising returned a profit of $10 million.

Given that the managers of these firms are likely to be risk-averse to some degree or another, and that there will be a significant lag before the firm can retaliate if the other firm were to increase its advertising, the focal firm will worry that if the other firm moves first to increase its advertising to $6 million then the focal firm’s profits will fall from $10 mil- lion to $5 million. The other firm will have exactly the same fear, so both firms are likely to increase their advertising expenditures to $6 million, and if they do they will end up in the lower-right-hand quadrant with profits down from the $10 million level to $8 million but this will be better than the worst outcome, which is to remain at $4 million advertising and see profits fall to $5 million.

The Prisoner’s Dilemma Problem and the Maximin Strategy In the previous example, the firms are subject to what has been called the prisoner’s dilemma. A prisoner’s dilemma is a situation in which two parties who act independently to make gains for themselves actually create a worse outcome for both parties. This is called the prisoners’ dilemma after the supposed situation in which two bank robbers are caught with a bag of money, which they tell the police they found lying in an alley. The police have no more than circumstantial evidence that these two men actually did rob the bank, but they are in possession of stolen goods, which would earn them a one-year prison sentence. The police put the men in separate rooms and interrogate them individu- ally. They tell each robber that if he confesses and implicates the other, he will walk free under a “state witness” deal, while the other will get eight years in jail. But if the other one confesses, the “state witness” deal is off the table and they will both get five years in jail. So the payoff matrix for the bank robbers looks like the one in Table 11.2.

Given the inability of the two prisoners to communicate with each other (and because “there is no honor among thieves”), each will be motivated to avoid the worst possible outcome, which is eight years in jail if he denies the robbery while the other robber con- fesses and implicates him. If each prisoner confesses, they each end up with five years in jail, which is far worse than the one year they would get if they both deny the robbery and are convicted of being in possession of stolen goods.

Table 11.2: The prisoner’s dilemma payoff matrix (jail sentences in years)

Options for Prisoner B

Deny Confess

Options for Prisoner A Deny 1 year; 1 year 8 years; 0 years

Confess 0 years; 8 years 5 years; 5 years

In each of the situations mentioned above, the players in the game are motivated to avoid the worst outcome. Each strategy has two outcomes that differ in value depending on

dou70192_11_c11_325-355.indd 347 11/1/12 5:03 PM

CHAPTER 11Summary

what the other player does at the same time—one of these outcomes is worse than the other. For example, if Prisoner A denies involvement in the robbery, the payoff will be either 1 year if Prisoner B also denies or 8 years if prisoner B confesses and implicates. On the other hand, if Prisoner A confesses, the payoff will be 0 years if Prisoner B denies or 5 years if Prisoner B also confesses. The worst outcome for each strategy is 1 year for denial and 0 years for confessing. The best of these worst outcomes is 0 years and so the best strategy for a person who dislikes jail is to confess. Choosing the option that has the best of the worst outcomes is called the maximin strategy, which signifies the maximum of the minimum outcomes.

Now let’s refer to the case of the mutual-dependence-recognized oligopolies choosing the level of their advertising budgets. If Firm A keeps advertising at $4 million, the profit payoff to Firm A is $10 million if Firm B also keeps advertising at $4 million, but profit falls to $5 million if Firm B raises its advertising to the $6 million level. Alternatively, if Firm A increases advertising to $6 million, the payoffs are $12 million if Firm B maintains advertising at $4 million, but only $8 million if Firm B also increases its advertising to $6 million. The worst of these payoffs is $5 million and $8 million, respectively. The best of these worst outcomes, $8 million, is likely to occur because the managers of both firms are risk-averse and prefer larger profits to smaller profits and will be motivated to follow the maximin strategy of increasing advertising to the higher level.

The prisoner’s dilemma problem applies to the oligopolists’ prices, advertising, product quality choices, and distribution system choices. Thus, they are at constant risk of reduc- ing their profits by cutting prices or launching nonprice strategies that actually reduce their profits because their rivals were motivated to do the same thing at the same time. These examples illustrate the incentive for oligopolists to make agreements to maintain prices and nonprice variables at current levels. Such agreements are known as price fixing, in the case of agreements to hold prices at current levels (or to raise them simulta- neously), or collusive agreements when two or more firms agree to maintain at current levels or increase or reduce other strategic variables simultaneously or at about the same time. As you will probably be aware, collusive agreements are illegal and are policed by the Antitrust Division of the U.S. Department of Justice. Firms can be fined millions of dollars for engaging in collusive behavior and the managers of such firms will also receive huge fines and jail terms as well. Clearly, when invited or tempted to collude with the managers of a rival firm—walk away and don’t even think about it! Even talking to man- agers of rival firms might be construed as circumstantial evidence of collusive behavior, so choose your friends and acquaintances carefully. Better to be subject to the prisoner’s dilemma problem out in the business world than to be subject to the problems of a pris- oner on the inside!

Summary

In this chapter we have examined nonprice competition, meaning the use of the firm’s strategic variables other than price, namely product quality, promotion, and place of sale. Each of these nonprice strategic variables is a demand shifter causing the firm’s demand curve to shift outwards to the right, if augmented, or inwards to the left if diminished. Nonprice competition essentially focuses on product differentiation and thus has no

dou70192_11_c11_325-355.indd 348 11/1/12 5:03 PM

CHAPTER 11Summary

application to pure competition where products are identical. Nonprice competition is an important adjunct to price competition in monopolistic competition, oligopolies, and even monopolies, since the monopolist can attract demand away from indirect competitors. The manager’s problem is to adjust the nonprice variables such that the firm’s product is seen as the superior value proposition by its target customer, where value is equal to qual- ity over price, and quality is broadly defined to include aspects of all three nonprice strate- gic variables. Using the value proposition approach allows us to note that some nonprice competition raises the value proposition by reducing the total price that the customer has to pay. For example, informative advertising reduces the customer’s search costs, and opening additional sales outlets reduces the customer’s transactions costs.

We saw that there will be diminishing returns to the augmentation of any one nonprice variable and, thus, the optimal level of that variable needs to be found. The usual profit- maximizing condition, that the marginal cost associated with the last unit sold should be no greater than the marginal revenue received from the sale of that last unit, is again uti- lized, but we noted that the marginal cost now includes not only production costs but also the incremental cost of product quality augmentation, promotion, and distribution. The marginal revenue from the last unit sold comes not from a movement down a demand curve but from a shift outward of the firm’s demand curve.

Thus, as nonprice variables are changed, both the demand curves and the cost curves will shift outwards and upwards, respectively. We utilized a new analytical technique, involv- ing locus curves, to show the path traced by the profit-maximizing price (i.e., average revenue) in each quality scenario and similarly, we used a locus curve to trace to average selling costs associated with each profit-maximizing output level. Introducing curves that are marginal to each of these average locus curves allowed us to find the output level where the marginal equality condition was satisfied, and this in turn identified the opti- mal level of the price and nonprice strategic variables, determined simultaneously.

In the real world, information search costs would typically make it economically ineffi- cient to implement the theoretical solution. But, understanding the process involved will assist the managers in utilizing whatever information they can obtain at reasonable costs. We noted that if records are kept of monthly sales levels and advertising levels, then corre- lation analysis can be conducted to ascertain the apparent marginal impact of advertising on sales. By utilizing dummy variables to indicate the periods before and after a product improvement is introduced, or a new distribution channel is opened, regression analysis can be utilized to estimate the impact of the change in product quality or the distribution system.

Finally, we focused on the mutual-dependence-recognized problem of the oligopolist, who must expect rivals to react to both its price and nonprice competition initiatives. Indeed, since nonprice competition needs to be premeditated due to the inevitable lags between deciding to take action and implementing action, oligopolists try to “steal a march” on their rivals by preemptively introducing nonprice initiatives using the maximin decision criterion. As a result, they fall foul of the prisoners’ dilemma problem that occurs when parties have conflicting interests and are not able to effectively communicate with each other to ensure that no one is taking self-serving action that will damage the other parties.

dou70192_11_c11_325-355.indd 349 11/1/12 5:03 PM

CHAPTER 11Decision Problems

Questions for Review and Discussion

1. It is sometimes said nonprice competition is hard to do well, whereas any fool can cut prices. Discuss this in the context of an oligopoly.

2. Why should we expect diminishing returns to apply to the effectiveness of advertis- ing and promotional activities? Outline several reasons.

3. Suppose that it would be possible to improve the quality of a product further and further until it was perfect. Why would the profit-maximizing firm not want to do that?

4. What is the rule for finding the profit-maximizing level of advertising expenditures in the short run, given that price and average variable cost levels would remain constant?

5. How is that profit-maximizing rule modified if (a) average variable costs rise as quantity demanded increases, and (b) if the price declines as additional units of the product are placed into the market?

6. Outline the logic behind the process for the simultaneous determination of the profit- maximizing price and quality levels.

7. Why is price elasticity of demand likely to be relatively high for search goods, as compared to experience goods?

8. In what ways has the Internet acted to increase competition among rival firms and raise the value proposition that the firm can offer to a customer?

9. Does the Internet have an impact on the advertising elasticity of demand, relative to pre-Internet times? Please explain.

10. Why do oligopolists face a prisoner’s dilemma problem when it comes to deciding on the level of their advertising budgets?

Decision Problems

1. The Thompson Textile Company has asked you for advice as to the optimality of its advertising policy with respect to its Product X. The following data is supplied. Assume that price would remain unchanged if advertising expenditure is varied, and that the marginal cost of production is constant at that level.

Sales (units) per month Advertising budget for Product X per month Advertising elasticity Price per unit Marginal cost of production

282,500 $56,000 2.5 $2 $1

a. Is Thompson’s advertising budget profit-maximizing at the current level? b. If not, can you say how much more or less it should spend on advertising? c. Outline the qualifications you would want to attach to your advice.

2. The McWilliams Beverage Company (MBC) manufactures and markets its own brand of bottled water. MBC’s present advertising and sales levels (month 9) are $4,000 and 99,500 bottles sold. The contribution margin is constant at $0.22 per bottle. MBC sells most of its product through vending machines in stores, bus stations, train stations, and other public places. Because vending machines accept only nick- els, dimes, and quarters, plus dollar bills, and must give change in coins, the cost of changing the price from the current level ($1.50) is almost prohibitive, so MBC

dou70192_11_c11_325-355.indd 350 11/1/12 5:03 PM

CHAPTER 11Decision Problems

engages mostly in nonprice competition, arguing that its water is more pure than other brands, on billboards, letter-box leaflets, and posters at the place of sale. MBC has noticed that quantity demanded responds to variations in the level of advertising and promotional expenditures. The firm has kept the following records over the past 8 months.

Month Sales (bottles) per month Advertising spent ($) per month

1 96,000 $3,400

2 103,000 $4,350

3 93,000 $3,750

4 111,000 $5,900

5 90,000 $2,600

6 76,000 $1,850

7 104,000 $5,200

8 120,000 $7,300

a. Plot the sales data against the advertising expenditures and sketch in what appears to be the line of best fit to the data.

b. Advise MBC as to the estimated optimal level of its advertising expenditures per month. Explain and defend your recommendation.

3. Flintrock Fixtures is a small firm that produces a variety of polished granite and marble kitchen and bathroom counter tops. The market for these items in the region is not highly competitive, since rival firms tend to focus on different segments of the market for new and renovated bathrooms and kitchens. Over the past few years, Charles Flint, one of the senior partners, has been experimenting with the levels of expenditures on advertising to determine the impact of this strategic variable on quantity demanded. Regressing monthly sales revenue against advertising expendi- tures, he has obtained the following regression equation: TR 5 110,482.5 1 2,318.6A – 103.2A2, where TR is monthly sales revenue, and A is advertising expenditures in thousands of dollars. This equation was derived from advertising data ranging from $1,000 to $8,500 per month and has R2 5 0.99, significant at the 1% level.

The present level of advertising is $6,000 per month, and Mr. Flint, who is more con- cerned with longer term profit and wants to maximize sales revenue in the short run, says he wants to increase advertising to $7,500 per month. Rocky Spinelli, the other senior partner, is more concerned with maximizing profit in the short run. He argues that given their pricing policy of marking up AVC by 100%, monthly profits would be increased by reducing advertising below the current level by at least $2,000 per month.

a. Assuming this data is reliable, what level of advertising would maximize sales revenue in the short run? Please explain.

b. What level of advertising would maximize profit in the short run? Please explain.

dou70192_11_c11_325-355.indd 351 11/1/12 5:03 PM

CHAPTER 11Decision Problems

c. Make an argument to support Mr. Flint’s position that a lower price in the cur- rent period will lead to greater profits over the longer term.

4. Vincenzo Pizzeria operates the only pizza parlor in a small western town, although there are several other fast-food outlets that are in competition with Vincenzo. The owner-manager, Vincenzo Fiorelli, feels that selling more pizzas is simply a matter of convincing people to eat out more often. Consequently, he holds price constant and advertises in the local newspaper and on the local television station. His pizzas come in three sizes with a variety of toppings, from plain cheese and tomato all the way up to the top-of-the-line deluxe pizza.

Vincenzo’s son Paolo has recently completed his business degree and has joined the family business as marketing manager. Paolo is interested in maximizing profits since his father has said Paolo can have half of any additional profit generated as a bonus. Paolo decides to conduct an analysis of the cost and demand conditions facing the firm. First, he examines the cost structure. Given the three sizes of pizza and the various combinations of toppings they are effectively offering a very broad product line. Paolo’s first task is to convert all the different pizzas into the terms of a common denominator, which he calls the medium pizza equivalent (MPE). Thus, a medium deluxe pizza is equal to 1 MPE, a small deluxe is equal to 0.75 MPE, and a large deluxe pizza is equal to 1.5 MPEs, with lesser weights in each size for lesser toppings (e.g., a plain tomato and cheese in the small size is 0.5 MPE). Paolo finds that the average variable cost of an MPE is $2.65 and that this is constant across a wide range of output levels. The first major decision Paolo makes is to standardize prices on all pizzas by marking up the AVC by 50%.

Paolo then conducts marketing research by interviewing a random sample of 500 customers and potential customers, and derives the following regression equation:

Q 5 28,105.1 – 5,842.2P 1 1,061.6A – 22.5A2

Where Q is the number of MPEs demanded per month, P is the price of an MPE in dollars, and A is the advertising expenditures per month. At present, prices are determined by a 50% markup on AVC and advertising expenditure is running at $8,000 per month.

a. Using graphical analysis, find the optimal price and monthly advertising levels. b. Confirm your graphical estimates algebraically. c. How much will Paolo’s monthly bonus be? d. State all assumptions and qualifications that underlie your answers.

5. The Silk Purse Cosmetics Company operates in close competition with several other major suppliers of cosmetics and toiletries. In this market, consumers do not seem to be very price conscious—if they believe a product can help them, they tend to buy the product, as long as it seems like a reasonable value proposition. Consequently, Silk Purse and its rivals tend to compete on the basis of claimed quality features, such as eliminating wrinkles and preventing hair from going gray. Silk Purse’s advertising budget is currently $25 million per annum, and it estimates that its rivals collectively spend $100 million annually. Silk Purse’s net profit for this year is expected to be $2.8 million. The vice-president of finance is worried that this profit

dou70192_11_c11_325-355.indd 352 11/1/12 5:03 PM

CHAPTER 11Key Terms

level will be insufficient to support continuation of Silk Purse’s research and devel- opment program, not to mention that dividends, taxes and management bonuses must also be paid from these profits. She suggests that a reduction of advertising to around $20 million would cause the profit situation to improve. The vice-president of marketing disagrees; he thinks that a reduction in advertising to $20 million will cause sales to drop by $10 million and profit to fall to $300,000. He advocates that Silk Purse should increase advertising to $30 million, which would increase sales by $8.5 million and generate profits of $4 million. The president, Mr. Hogg, feels that rivals are likely to react to protect their market shares.

a. Construct a payoff matrix and explain the vice-president of finance’s argument to Mr. Hogg.

b. Explain the vice-president of marketing’s argument to Mr. Hogg. c. What information do you suggest Silk Purse gather before making any decision

to change its level of advertising?

Key Terms

average quality cost (AQC) The total costs of quality at a particular level of quality, divided by the number of units of output subsequently produced and sold.

collusive agreements Illegal agreements, whereby two or more firms agree to set or maintain prices or other strategic variables at particular levels, or to change the value of strategic variables simultaneously or at about the same time.

credence good An experience good for which the quality is both unobservable before purchase, and where the quality is likely to vary from the previous times that customers have purchased that product, such that the consumer must give cre- dence (believability) to the quality claims of the seller.

Internet promotion A form of company advertising that includes informative and persuasive material being posted on websites, as well as purchasing Google “ad-words” and paid advertisements that pop-up when people search for specific information on the Internet.

just-in-case A philosophy of maintain- ing inventory or another resource at a relatively high level in case it is suddenly demanded by consumers or needed by the firm, to avoid losing business due to not being able to supply products to consumers.

just-in-time An approach to stocking components or finished goods inventories where a firm maintains a relatively small supply of these, but is able to supply products quickly when a buyer wants to purchase a given item.

locus of average quality costs (LAQC) A hypothetical line joining combinations of the average quality cost and the output level, at each profit-maximizing output level, for different quality scenarios.

locus of marginal quality costs (LMQC) A line joining the combinations of marginal cost of quality augmentation and the output level, at each profit- maximizing output level, for different quality scenarios.

dou70192_11_c11_325-355.indd 353 11/1/12 5:03 PM

CHAPTER 11Key Terms

locus of optimal marginal revenue (LOMR) A line joining the combinations of the marginal revenue associated with quality augmentation and the output level, at each profit-maximizing output level, for different quality scenarios.

locus of optimal prices (LOP) A line joining the profit-maximizing price and output coordinates for different quality scenarios.

marginal-equality condition The require- ment that the marginal revenue obtained from the adjustment of a strategic variable is just equal to the marginal cost associ- ated with that adjustment.

market research A process where the firm investigates the propensity of potential consumers to actually purchase the firm’s product, which may include interviews, focus groups, and surveys to estimate the likelihood that consumers will purchase the product.

price fixing The illegal process of setting the price of a service or good in collusion with other firms, and thereby setting a higher price than might otherwise occur if the firms behaved competitively.

price war A situation of aggressive price competition where competing firms keep lowering their prices to undercut the prices set by other firms in the market.

prisoner’s dilemma The situation where the strategy that maximizes the payoff to an individual or firm actually makes that individual or firm worse off because of the simultaneous choice of the same strategy by other player(s) in the contest, or game.

zero-sum game A rivalrous situation where the loser’s damages are of the same absolute magnitude as the winner’s gains, such as the change in market share when two firms compete with each other.

dou70192_11_c11_325-355.indd 354 11/1/12 5:03 PM