Market power on proprietary aftermarkets

Market power on proprietary aftermarkets

An insight into the printer manufacturer’s ability to freely set prices in their aftermarkets

Market power on proprietary aftermarkets

An insight into the printer manufacturer’s ability to freely set prices in their aftermarkets

Preface

This document is the result of exhaustive research to complete the project for my master thesis. It concludes the final stage of my study Business Economics at the faculty of Economics at the Rijks universiteit Groningen.

From the outset, this research project first sought to identify and explain the most important variables that shape the market for laser beam printers and, based on those market characteristics, create a market forecast model. The econometric model serves to explain market dynamics, and is ultimately used to make forecasts of the market. However, due to its unforeseen complexity, solving the model would require a level of expertise in the field of Econometrics that went beyond (Business) Economics.

Subsequently, this master thesis is a spin-off of the initial research focussing on perhaps the most interesting part of the laser beam printer market: the aftermarket for consumables and particularly supplier aftermarket market power. Furthermore, if this business is structured properly, it can prove to be a steady profit stream for companies even when their main products face heavy competition ultimately generating little profit.

This research would not have been possible without cooperation of Konica Minolta Printing Solutions Europe B.V, which was my employer when I wrote most of this thesis. This organisation was generous enough to provide me with access to all data available on printer markets. I would like to emphasise that opinions or conclusions in this document do not necessarily reflect those of Konica Minolta PSE nor does it necessarily influence company policy or strategy.

I would like to thank friends, colleagues and fellow students for their support and feedback. Also I would like to thank dr. H.F. Lanting and dr. D. Wiersma, both Factulty of Economics Rijks universiteit Groningen, for their assistance and valuable input.

Summary

In today’s business environment, companies are faced with an ever increasing competitive marketplace. Therefore, in order for companies to stay ahead they are required to continuously revise their business models to meet fluctuating market conditions. In this research, I have focussed on the Laser Beam Printer market to illustrate this concept. Printer vendors currently find themselves in a crowded market of increasingly tough competition. For example, the German monochrome laser printer market (for printers with a print speeds up to 35 pages per minute) which has been utilised for this research has 40 competitors. In a competitive market, companies are unable to earn a margin over marginal costs and are forced to explore ways to find more profitable markets.

To combat this, Laser printer manufacturers have developed very profitable after markets for toner cartridges. This research sets out to establish whether a company has any market power on its proprietary aftermarket.

The research model that has been presented, suggests that aftermarket power can be evaluated by comparing retail prices of aftermarket products to fore market products. To accomplish this, however, four conditions have to be met in order to establish whether or not changing aftermarket prices will cause a negative market response in the aftermarket or the fore market.

Contents

Preface ... 3

Summary ... 4

Contents... 5

1. Introduction... 7

1.1 The application of a new business models ... 8

1.2 Management dilemma ... 9 1.3 Management question... 10 1.4 Research question ... 11 1.5 Research lay-out ... 11 2. Literary review... 13 2.1 Markets ... 13 2.1.1 Market clearing... 15 2.1.2 Arbitrage... 16 2.2 Competition ... 17

2.2.1 Perfect competition vs. monopoly ... 17

2.2.2 Exercising market power ... 19

2.2.3 Pricing strategies... 20

2.3 Foremarket and aftermarket dynamics ... 21

2.3.1 Link between the fore market and the aftermarket ... 22

2.3.2 The foremarket ... 22

2.3.3 The aftermarket... 23

2.4 Market power in the proprietary aftermarket... 24

2.4.1 Link between fore and aftermarket... 24

2.4.3 Price elasticity... 25

2.5 Price level development in after markets... 26

2.6 Conclusion... 27

3. Research methodology... 28

3.1 Research model – conditions... 30

3.2 Aftermarket prices ... 30

3.3 Limitations of the research model ... 30

4. Findings and conclusions... 32

4.1 The market... 32

4.2 Testing the conditions... 35

4.2.1 Foremarket competition... 35

4.2.2 Link fore market vs. aftermarket ... 38

4.2.3 Third party supply in the aftermarket ... 38

4.2.5 Price elasticity... 41

4.3 Aftermarket power... 44

5. Conclusion and further research ... 46

5.1 Evaluating the research model... 46

5.2 Reviewing management dilemma and management question... 47

5.3 Further research ... 48

Appendix I – Eurostat: Harmonised annual average consumer price indices... 50

1. Introduction

Competition is present in virtually any marketplace both in Western countries and in developing countries around the world. It is caused by companies competing in a marketplace (with a similar

product or a closely related substitute product) for the grace of the consumer1. A market is

created through an interaction of buyers and sellers. The actual buying and selling and even the intention to buy or sell creates a market and subsequently, market prices.

How competition is translated to real life situations is subject to continuous change. Furthemore, one could argue that companies create competition by themselves just by being in the marketplace which as a result causes rival companies to respond. Additionally, competition is also caused by the consumer. Each customer searches for the ideal product given his own set of selection criteria. These criteria can be related to the product itself (e.g. quality, price, value for money) or external factors such as fashion or the influence of other products. Assuming there are several suppliers in a market, every company tries to outperform their competition by offering the best product to persuade the consumer to buy. First this will probably lead to more differentiated or advanced products. Increased variety in the marketplace causes the consumer to raise his expectations from the market, due to for example, availability of more advanced products. The consumer remains in the state of constantly having to adjust his set of selection criteria. A second reaction to increased competition from suppliers is to lower prices as competition increases.

Adding to this vicious circle is the availability of more transparent information about markets to potential consumers and companies. Increased knowledge provides both parties with a better view of the marketplace, which ultimately accelerates changes in the marketplace.2

1

From of course a vendor or supplier point of view.

2

1.1

The application of a new business models

Companies are always looking for more advanced ways of setting up their business models in order to better cope with the competition and the marketplace. Especially in fast changing markets, a company’s level of adaptability to market changes is crucial to their survival. Numerous business models have been developed in the past to cope with an ever changing environment. The goal is to increase margin on the cost of a product, or at least to devise ways to prevent margins from decreasing. This is crucial to create a healthy financial situation for the company in the long term3.

For the purpose of this research focus will be limited to Hardcopy peripherals, specifically Laser Beam Printers. The market for laser beam printers (LBPs) has become increasingly competitive in the last decade, especially the last 5 years. There are two causes for this trend. Firstly, the range of products in this market have become very homogenous (with regards to technical specification). The trend towards homogenous products is driven by the nature of demand. It is estimated that due to the automation and digitisation of virtually every office in Western Europe, around 90% of printer-hardware is used to print daily office documents4. These activities do not require any special or extraordinary features from the hardware. The printer is used to make one or a limited number of printouts of a document as it is displayed on the user’s screen. In general, a laser beam printer is not used for sophisticated printing, as for example digital printing, or high volume printing. For these kinds of activities, specialised hardware exists.

In addition to this impact from the demand side, the trend towards homogenous products is accelerated from the supply side. The technical expertise of the vendors in this market has developed to a point where every vendor is capable of designing and producing good quality, low cost printers. In case a vendor does not have a competitive product in a certain segment of the market, it is not unusual to source a product or part of a product5 from other vendors to add to its own product line-up. This results in a rich choice of almost the same printers where only marketing message and brand name are the real differentiators.

3

For the sake of argument assuming that as long as revenues are higher than associated costs, a company’s financial situation is considered to be healthy.

4

This is a generally accepted assumption in this industry, however not empirically tested.

5

In addition to the fact that the market becomes more homogenous, increased competition in the market is caused by the status of maturity in that market. Especially in Western Europe, the market for laser printers is close to maturity. The demand for printer hardware is slowly growing into a replacement demand. Second to that, in general printer hardware has a longer economic lifecycle causing end users to replace printers less often. Measured in quantities, demand is therefore flattening and vendors are competing in an environment of diminishing demand.

Increased competition in the LBP market has resulted in lower margins on hardware to a point of almost non existent. Companies are forced to sell their products at cost. As mentioned, it is in the nature of a company to adjust its business model to market circumstances. Trends in the printer market also force printer vendors to reconsider their business models. Vendors initially focussed on selling hardware (printers) to end-users because there was no economic or financial incentive to do otherwise. While the market for hardware becomes more competitive and less profitable by the day, the market for consumables (mainly toner cartridges) and hardware options (hardware add-ons to the printer) are still very profitable. Hardware options are usually used for product differentiation. Consumables, however, are sold in after markets. The focus in the business model, therefore, causes a shift towards more profitable after markets for the laser beam printer aftermarket products.

1.2 Management

dilemma

In the case that the aftermarket is characterised by competition or even perfect competition and not a monopoly, a vendor is limited in setting prices to a level that best suits the financial goals of the company. In the worst case scenario, the company would be forced to adjust its prices to the market. If the vendor has a monopoly (or at least strong market power), it would suggest that a vendor can maximize its profitability by simply setting prices so that Marginal Revenue equals Marginal Costs [Martin 2001]6.

Furthermore, it for example also suggests that consumers do not consider aftermarket economics when purchasing a printer. However, consumers are aware of the fact that a printer requires consumables in order to stay functional. Therefore, the link between the aftermarket and its foremarket should remain. It is however not clear to what extend aftermarket characteristics influence the fore market. The fact that prices in the aftermarket do not significantly decrease over time, would suggest there is no perfect competition in the aftermarket. However, this does not mean the vendor has the luxury of monopolistic power in the aftermarket.

For the vendor a management dilemma is the fact of not knowing whether he possesses significant (or even monopolistic) market power on its proprietary aftermarket for consumables. Significant market power would enable the printer vendor to set prices (within certain boundaries) without causing considerable negative market impact in either its hardware or aftermarket revenue.

1.3 Management

question

Based on the management dilemma as stated before, a management question can be derived. The management question for the purpose of this research formulated as:

Is it possible to setup a profitable aftermarket to increase overall company financial performance, without compromising our current market position in the market for laser printer or any printer related products?

6

1.4 Research

question

The potential negative market impact as described before is in fact two fold. The first is from the consumers. This would be caused by consumers taking aftermarket circumstances into consideration when purchasing a printer. If the aftermarket price becomes too high, this will greatly impact printer sales. Furthermore, this is especially the case when customers take into account the total lifetime value or costs when making their printer purchase.

Secondly, the vendor can expect competition to react swiftly with a price adjustment (competitive behaviour). A vendor can, for example, choose to sell products into another vendor’s aftermarket. It is, however, more likely that third party suppliers will copy the design of the original products and sell them into the vendor’s aftermarket.

This thesis does not aim to research the possible market response itself, but mainly focuses on the question if a vendor is able to set prices without triggering any kind of negative market response.

Consequently, the research question is:

Is a Laser Beam Printer vendor able to exercise sufficient market power in its proprietary aftermarket for consumables enabling it to freely set prices that would positively impact is financial performance, without causing a significant negative market response?

1.5 Research

lay-out

This research is an exploratory study. It attempts to establish to which extent a vendor has some kind of market power on its proprietary aftermarkets. To find an answer to this question, this thesis is set up in three stages.

factors that determine if a company can execute market power and the factors that might prevent a company from doing so. When the scope of the research is set, a research model will be developed based on the available literary research. The research model is to help answer the question raised in the previous paragraph. If possible a method to empirically test the impact of the factors will be found.

The second section will focus on the empirical research. The empirical research will be carried out based on a set of market data which describes the market for laser beam printers and the market for consumables (mainly toner cartridges) for these printers. The dataset consists of a set of time series, all secondary data. The empirical research aims to quantify the factors that might determine to which extent a company can execute market power.

2. Literary

review

This chapter will set a framework for the empirical research. First a number of concepts are defined and the boundaries for this research are examined. Based on this framework, competitive markets versus less competitive markets (and monopoly) are deliberated. In conclusion, this chapter will suggest a strategy for price setting (which is relevant to market power), and set a model for price setting in the aftermarkets.

2.1 Markets

For the purpose of this research, the concept market needs to be first defined. One of the important characteristics of a market is where goods or services are traded at a certain price and the trading of goods itself (supply versus demand) over time determines the market price. With this in mind, the definition by Pindyck et al. covers this characteristic:

A market is a collection of buyers and sellers that, though their actual or potential interactions, determine the price of a product or set of products. [Pindyck et al. 2001]

This definition concludes that prices are set through the mechanism of supply and demand. It also suggests that prices are not only linked to the actual trading of the goods, but also by the intention of the market participants to trade goods. This is a valid assumption, which has is also relevant for the the printer hardware market and consequently for this research. When a printer vendor introduces a new printer model to the market, it sets a price so that the product is at least competitive with the most expensive model currently available in the market or with the most popular product available. More often, however, prices are set at the level of future competitor products, anticipating the launch of these products.

in the printer market and sold at a price level which is higher than the base model. In addition to technical features, the products also vary on price, which helps the customer to easily recognise the variations between products. Often consumers accept a price gap between differentiated products which is higher than the additional production costs would require. Price differentiation is however mainly used as a marketing tool rather than as a way to earn back production costs. All differentiated products are, however sold in the same market, with the same suppliers and customers. The definition therefore needs to allow for different price points for one and the same market.

A second issue not covered in the definition above is the geographic boundaries of a market (as also recognized by Pindyck et al.). It is not very likely that an end user travels great distances or spends much time to locate the cheapest printer available7. Since mass markets such as the printer market become more transparent, it is however more likely that end users will use, for example, the internet to order printers from another supplier than their local IT hardware store. Geographic boundaries can, therefore, play a role when discussing the printer market. A definition which covers multiple price points and the geographic boundaries is for example:

A market is a region within which a group of varieties (at least slightly differentiated products) for which prices tend to equality, adjusting prices for differences in cost of supply and for differences in product characteristics. [Martin, 2001]

This definition also allows for price adjustments due to differences in the costs of supply. Within this research costs of supply, however, do not play an important role for a printer vendor since almost all distribution is outsourced. Consequently, the costs of supply are minimised by the efficiency of the market.

Since both definitions contain elements that are important to describe a printer market, both are combined to one definition which shall be used for this research:

A market is a combination of buyers and sellers in a specific region that, though their actual or potential interactions, determine the price of a product or set of products, while adjusting prices for differences in product characteristics.

7

2.1.1 Market

clearing

Within the definition it is stated that a market price is determined and even more than one price in a market with several differentiated products. The question remains how a price level (or multiple price levels in the case of differentiated products) and a matching output quantity in a market is established. For now, a market with one homogenous product is assumed, however the theory remains the same for a market with differentiated products.

Theory states that supply to a market has a positive relation with price, i.e. when the price increases, companies will increase their supply to the market. If market output is plotted against all market prices, the supply curve is created. The shape of this supply curve is determined by the costs of producing the product and marginal costs in particular [Besanko et al., 2000]. A supplier will make a margin and stay in a market as long as the total revenue it receives for its products is higher than the total costs spent to produce the products – considering that the price is higher than the average costs per product.

The demand curve assumes a negative relation8 with the price of a product which means that if

the price of a product increases, the demand measured in quantity will decrease [Besanko et al., 2000]. Demand is influenced by much more variables than just price. For example, price of related products as well as marketing and promotion for products and of course, as stated before, consumer expectations. For the sake of simplicity in the model, all variables other than the price of the product are assumed to remain fixed. If market demand is plotted against prices, the demand curve is created.

The shape of the demand curve is determined by the price elasticity of demand [Besanko et al., 2000]. If the quantity demanded varies highly with only a small variance in price, price elasticity is considered to be high, and vise versa. If price elasticity is high, the demand curve will tend to move more horizontal. In contrast, if price elasticity is low the demand curve will be more vertical, i.e. a variance in price will lead to only a small variance in quantity demanded. Price elasticity can have several causes. If for example is it easy for consumers to find a substitute product providing the opportunity to switch to another product, it is more likely that they will switch when the price for the initial product increases. The quantity demanded for this product will, therefore, decrease more rapidly than if there is no easy access to a substitute product.

8

Another example, which decreases price elasticity is when the consumer has no choice but to buy the product (consumer is locked in). This can be the case when significant costs are incurred when a consumer wants to switch to another product.

Since both the demand and supply curve have an opposite relation with price, the curves cross. This equilibrium is the combination of price and quantity where the supply (in quantities) matches demand (in quantities) [Pindyck et al., 2001]. The market price is determined via market clearing, i.e. both price and quantity keep on varying until they match the equilibrium price and quantity. If supply exceeds demand, a market surplus is created. Suppliers will lower their output and drop the price. Due to the lower price, demand for the product will increase until the market clears again. The opposite will occur if demand for the product exceeds supply and a market shortage is created. The suppliers will react by increasing prices and output due to the increased demand, until the market clears and the market reaches the equilibrium price and quantity. The market equilibrium will not change until there is a shift in supply or demand, which will change the shape of either one or both curves.

2.1.2 Arbitrage

The market clearing process as described above assumes trading of goods within a market. As the definition of a market earlier in this chapter suggests, there are physical boundaries limiting the size of every market. It is, however, very well possible to ship goods from one market, for example a market in France, to another similar market, in Spain for example – should prices in both markets permit. The goal is to buy products in one market at market price and sell those products on another market where the market price is higher. This process of arbitrage itself prevents large variances in prices in otherwise separated markets, and therefore again limits arbitrage. In the printer market this phenomenon also occurs, however is very limited because printer vendors usually operate globally and can, therefore, keep prices roughly similar in markets where arbitrage can occur9. For the sake of this research, it is therefore assumed that there is no in- and export between markets once vendors have sold their products into the market.

9

2.2 Competition

Now that is established how in a market the price and output quantity are determined, the next question is how companies interact with each other. The main variable that determines the behaviour of a company lies in the competitive landscape of the market. At one end of the spectrum is the monopoly with no other competitors, and perfect competition with virtually unlimited number of competitors at the other end of the spectrum.

2.2.1 Perfect competition vs. monopoly

Companies try to maximise their profits. They will continue this behaviour in whatever market they are in. Suppose a company is in a market characterised by perfect competition, therefore experiencing competition from a number of other companies. A perfect competitive market assumes four basics [Pindyck et al., 2001; Martin, 1994]. First, all companies in the market have to take the market price as given. This links back into the assumption that actions of an individual company will not influence market price significantly, because none of the competing firms have a market share large enough to influence the price. It also assumes that companies do not collude. If comparable behaviour between companies is observed this is assumed to be a similar reaction to changed market circumstances.

Second, all products in the market are perfectly substitutable and therefore homogeneous. This of course, is in close relation with the first condition: if a company sells the same product as its competition, it cannot raise prices. If it does choose to increase prices, potential customers will purchase a competing product. As discussed before, this assumption also holds true even if products are slightly differentiated, because this will be ultimately corrected by small price variances.

Fourth, there is complete and perfect knowledge. This means that all buyers and sellers have the information that is relevant to them, for example information about available technology for sellers and information about price to buyers [Martin, 1994].

In addition to the four points presented above, Martin [2001] also states that there have to be many small buyers and sellers. The size “small” is up for further debate. Therefore, a company is considered small as long as it does not have a market share large enough to significantly influence market price [Martin 1993]. Since this assumption on price taking also exists in the first condition as described above, the small firm condition is not mentioned separately.

As I mentioned before, a company strives to maximise its profit. It does so by setting a price and quantity so that its marginal costs equal its marginal revenue. As long as the price is higher than the average costs for production, the company will make a profit. In a perfectly competitive market it is assumed that there is free entry and exit. When companies in the market are making a profit (i.e. the price is higher than the average costs) there will be an incentive for other companies to enter the market as well. If the new company decides to enter the market, the total quantity of goods supplied to the market will increase, forcing the supply curve to change. The increased number of companies and their increased supply to the market force the supply curve to shift to the right, which resembles the higher quantity supplied. However, the increased number of companies does not change the shape of the demand curve. The combination of fixed demand and changed supply has the effect that the market price will drop until the market clears and is in its equilibrium again.

For the individual company in the market this does not have a positive effect on its economics. The output for one company decreases as it attempts to minimize its average costs per product. The price on the other hand has also decreased (due to increased supply) and therefore its revenue, causing its profits to decline (delta between price and average costs). For the individual company, the market is attractive until its profit turns into a loss. Until that point the company will stay in the market. Furthermore, until that moment the market will also attract new entrants. At the point of breakeven, where price equals average costs per product, the trigger for new companies to enter the market will disappear and new companies will not enter the market. At this point, there is a perfect competitive market, where all companies have to accept the market price as a given.

the competition, because in a monopoly competition doesn’t exist. The only thing the supplying company needs to worry about, if it is to maximise its profits, is the quantity demanded at a certain price. The company will maximise its profits by equalling its marginal revenue with its marginal costs. Therefore, the company needs to determine its own cost structure, and the shape of the demand curve [Pindyck et al., 2001]. Besides the freedom to set prices, a second condition for a monopoly is that entry is blocked [Martin 1993]. While the first assumption refers to the fact that the monopolist currently faces no competition, the second condition refers to the fact that the monopolist will not be facing any competition in the future either.

More interesting than how a company can maximise its profits or block potential competition, is what the company can do with its position under certain circumstances.

2.2.2 Exercising market power

Market power can be described as the ability of buyer or seller to affect the market price of a product [Pindyck et al., 2001]. Several authors have discussed the welfare and economic impact of monopolies and market power and how public policy can limit any potential negative effects10. The reoccurring topics are the lost consumer surplus and deadweight welfare loss to address market inefficiencies of monopolistic markets [Martin, 1994]. This research, however, aims to discuss this topic from the point of view of a supplier. Although it is acknowledged that monopolies or markets with limited competition are less efficient, the topic will not be further addressed.

10

See for example any book on micro or industrials economics (e.g. Pindyck et al., 2001; Martin, 1994, 2001; Besanko et al., 2000) and also several, some rather interesting articles (e.g. Ho et al., 2003; Paarlberg

2.2.3 Pricing

strategies

If it is assumed that a company has a significant amount of market power, it has the ability to set

prices and to develop a pricing strategy11. Also assuming that the company is aware of the

workings of supply and demand12, and the nature of the demand curve in specific, it is able to fit the pricing strategy to the markets it operates in.

This section focuses on price discrimination as it is the basis for virtually all pricing strategies in the laser printer market. The practice of price discrimination lies in the fact that at all prices charged (within reasonable boundaries) customers can be found. As mentioned before, the demand curve is negatively related to the price, i.e. at a lower price, demand in quantities will increase. In contrast, if price increased demand in quantities will decrease, but there will still be buyers willing to buy at the increased price level. Therefore, the company should be charging different prices to different consumers in order to optimize company performance (Pindyck et al., 2001). In other words, the company has the ability to capture the consumer surplus and deadweight loss that exist if only a single price point is used [Martin 1993]. The challenge is to identify the different groups of customers.

There are three basic forms of price discrimination. First degree price discrimination is the ideal situation where the company is able to charge the highest price to each customer that is willing to pay (the customers reservation price [Pindyck et al., 2001]). This technique is however undermined by two things. First, if a buyer knows that other buyers are paying a lower price, that customer will not be willing to pay the higher price. Second is arbitrage (Martin, 1994), if a customer knows that other customers are willing to buy the product at a higher price, he is likely to resell his product to these customers. This form of price discrimination is therefore, only suitable for products for which the buyer and seller have to negotiate a price on an individual basis and which the buyer cannot sell to other customers.

Second degree price discrimination can be used when consumers purchase large amounts of the product. Different prices are charged for different quantities of the product (e.g. quantity

11

That is, a pricing strategy more advanced than the ‘let’s add a markup to these prices’ strategy, which seems to be the case for some companies when they find themselves in a situation where they can exercise market power.

12

discounts). All customers divide themselves in groups by choosing to purchase a certain amount, and therefore choosing a certain price point.

Third degree price discrimination case suggests that all customers are to be separated into two (or more) consumer groups by segmenting the market into two (or more) separate markets [Martin 1993]. Groups and markets are separated based on the difference in price elasticity and demand characteristics, therefore each group has their specific market demand curve [Martin, 1994; Pindyck et al., 2001]. The supplier is able to charge specific prices to the consumer in each group

based on the different demand curves13. An example of third degree price discrimination is

creating an aftermarket. A special case of aftermarket, and the one example on which this research will focus, is the proprietary aftermarket.

2.3

Foremarket and aftermarket dynamics

As discussed, the decreasing margins in the market for printers forced printer vendors to look for other ways to make a profit, hence a price discrimination model as discussed. The model, as discussed, requires the vendor to create an aftermarket.

The business model requires a market, which is split into a foremarket and an aftermarket. Suppliers in this situation generally try to gain a large aftermarket by creating an installed base in the foremarket. In this situation, margins in the foremarket are almost non-existent since many suppliers are active in this market, which is most appropriately described as a perfect competition. Revenues and profits in the aftermarket can be maximised, because the company has increased market power and therefore, is able to change prices. Some examples are the game console industry (game console itself is the foremarket product and the games are the aftermarket products, see also Garcia Marinoso 2001), post-paid mobile phone industry (fore market: initial payments (for the telephone), aftermarket: monthly payments) and, as one would expect, the laser printer industry (foremarket: printers, aftermarket: consumables).

13

2.3.1 Link between the fore market and the aftermarket

The foremarket printer and the aftermarket consumables form a system as mentioned by Garcia Marinoso (2001):

Systems are goods that consist of various components, which can be bought separately but must be used together.

In general, a customer buys one of the components in the system and continues to buy the other component in order to improve the system. A special case is where the supplier creates a system where both components need to be bought in order for the system to be functional. In this case, the supplier creates different life expectancies for both components. This implies that one of the components needs repurchasing in order for the system to stay functional. Doing so, a foremarket and aftermarket model is constructed where the customer is tied to the company for a longer period of time. In the foremarket, the long life product is sold which builds up an installed base (Klemperer, 1987). A short life product needs to be added to the installed base in order for the system to remain functional; the short life product is sold in the aftermarket. The model with products with a different life expectancy is more valuable to a company than the model where the system of products can only be improved.

The fore and aftermarket consist of two separated markets where consumers can be divided into two groups, consumers who did not yet buy the long life product (foremarket consumer group) and consumers who already did (aftermarket consumer group). This ties back to the third degree price discrimination model as discussed. Since there are two separated, but linked markets, the supplier can use the two to optimise company performance.

2.3.2 The

foremarket

The foremarket is used to create an installed base. In this foremarket, the supplier has to charge low margins, in fact accept market price, because in this market the company will experience competition from many competitors. All competitors will offer more or less the same product. In the case of the printer market, all products offered in a certain segment14 are homogenous with

14

respect to technical features. All products provide the end-user with the capability to print out digital documents in a certain quality at a certain speed. This segmentation in the foremarket in itself is also a form of price discrimination. This research, however, will primarily focus on the relation between a foremarket and its aftermarket.

2.3.3 The

aftermarket

As mentioned before, this research focuses on proprietary after markets. Other than in the foremarket, in the proprietary aftermarket the supplier tries to establish a monopoly. This is mainly due to the technical incompatibility suppliers create between foremarket components and aftermarket components of different brands (Farrell & Salone, 198615). This means that when a consumer has chosen one particular brand’s foremarket product it has to buy the aftermarket products of that brand in order for the system to be functional16. If a consumer wants to change to the aftermarket products of another supplier, it experiences switching costs: the consumer first has to buy the foremarket product of that supplier. Switching costs make sure that a consumer is locked into an aftermarket of a particular supplier and therefore gives the supplier the possibility of exercising market power in his aftermarket (Emch 2003, Corts 1998). As long as the costs for purchasing the aftermarket product do not exceed switching costs, the consumer has no incentive to switch from one supplier to the other.

15

Although this article deals with incompatibility between different evolutions of a technical standard, the theory discussed is also applicable to the case as described in this research.

16

2.4

Market power in the proprietary aftermarket

From the previous chapters, four topics have emerged when answering the question if a supplier can exercise market power on its proprietary aftermarket. The first is the way in which the aftermarket is linked with the foremarket. Second is the availability of third party products on the aftermarket, and their impact on the market structure. Third is the price elasticity of market demand. The final question is the price level development in the aftermarket over a longer period of time. The initial three topics are conditions which can prevent the company from having market power in the aftermarket. If these three topics, however, do not significantly impact the aftermarket, the fourth topic determines if the company in fact possesses market power.

2.4.1 Link between fore and aftermarket

When buying a foremarket product, it is obvious that current price and quality of the foremarket product are taken into consideration, but it is not clear to which extent the consumer also takes price and quality of any associated aftermarket products into account. Garcia Marinoso (2001) argues that the intensity of first period competition depends crucially on whether or not consumers rationally anticipate aftermarket prices. This is a result of the tendency of suppliers to exploit locked-in customers. If the supplier has a larger market share in the foremarket, this will lead to higher aftermarket prices, because of higher price mark-ups. When the consumer is aware of the dissimilarity in price mark-ups between suppliers in the aftermarket, he is more likely to take a greater consideration when selecting the foremarket product.

to investigate the market thoroughly for aftermarket conditions. Marketing campaigns stimulate this behaviour by emphasizing printer specifications rather that aftermarket costs (e.g. prices of toner cartridges). These marketing activities influence consumer preference operate similar to how consumer opinion is guided through confirmatory reasoning (see for example Chernev 2001). When purchasing a printer, a brand or product choice is made based on foremarket price and quality and general brand characteristics.

2.4.2 Third party suppliers

The impact of third party suppliers in an aftermarket have is immediately evident. If the number of companies in a market increases, the market will become more competitive. In an ideal situation, a company will experience no competition in its proprietary aftermarket. In this case, it has a monopoly and it can freely set prices that ensure maximum profit. In contrast, a perfect competitive market means that a company has to accept market price as a given.

Currently there are a number of third party manufacturers active in the European market. The question that remains: is how many are active, what is their market share in the aftermarket and if they limit market power.

2.4.3 Price

elasticity

Price elasticity determines to which extent the demand for a product will change as the supplier changes its pricing model [Besanko et al., 2000]. A low elasticity will give the firm more freedom to set its price; a price increase will not lead to a huge loss in demand.

An exception to this rule, is if the system of goods from the company becomes more expensive to use that the system from a rival brand. In this case, switching costs will not prohibit a customer from changing systems. In aftermarkets for printers, the price elasticity remains low until actual prices increase to the cost of changing brands.

2.5

Price level development in after markets

If the conditions as previously mentioned are all met, market power to set prices in an aftermarket can then be established. Under normal circumstances this would be done by testing if a company is charging a margin over marginal costs of a product. This is, however, not possible in the aftermarket for laser printer toner, because of two reasons. The first reason is that the production of toner is not a continuous process with a linear costs function. So it is not possible to match the exact costs of production with a specific group of toner cartridges as they are sold in the market. Therefore, it is also not possible to calculate marginal costs. The second reason, and more importantly, there is no data publicly available on the actual costs or marginal costs of toner (total costs including production, shipping and selling of the product). Printer vendors do not share these data.

2.6 Conclusion

3. Research

methodology

The research question as stated in chapter 1 raised the question whether a vendor is able to exercise sufficient market power in its proprietary aftermarket for consumables to set prices so that they positively impact is financial performance, without causing significant negative market response. The literary review highlighted that this can only be empirically tested if a number of conditions are met. First, one new condition is added.

1. Competitive foremarket. The theory as discussed only holds is the company experiences a highly competitive environment in its foremarket. This condition is set, because of the empirical test to prove market power.

From the previous paragraph, the following three conditional factors were identified:

2. The link between aftermarket and foremarket. There cannot be a strong link between aftermarket and fore market. It was suggested that if customers in the foremarket do take aftermarket circumstances into account when choosing a foremarket product. This limits the company to freely set its prices in the aftermarket.

3. Third party suppliers in the aftermarket. There cannot be any other suppliers in the aftermarket of the company with significant market power. If a market becomes more competitive, the company is limited in setting its own prices, because the company will have to follow market price.

4. Price elasticity. Price elasticity cannot be too high. High price elasticity will limit the company in setting its prices.

If the above four conditions are met, a company’s ability to set market prices can be empirically tested:

In the model below, the conditional factors as well as the empirical test to prove market power is graphically displayed.

Graph 3.1, Research model

Foremarket Aftermarket

2. Link foremarket

and aftermarket

1 Competitive

environment Potential non-competitive environment

4. Price elasticity

market demand

Are the following conditions met?

1. the foremarket is a competitive environment

2. the link between aftermarket and fore market has no

significant impact

3. third party suppliers do not have enough market power

to significantly influence market prices, the aftermarket can therefore considered to be non-competitive

4. price elasticity is not high enough to significantly limit

the companies freedom to choose prices.

3. Third party

supplier

Are prices in the aftermarket kept at the same level while prices in the fore market erode?

3.1

Research model – conditions

The first condition, a competitive foremarket will be tested via available market data. Market data about printer markets is available – both sales volumes and sales value in terms of a longer period. The second condition, the link between fore and aftermarket cannot be tested via for example market data. No empirical data with regard to the tested market is available. The condition will be theoretically discussed.

The third condition, third party suppliers in the proprietary aftermarket will be test via available market data. Same as for the foremarket data, this dataset contains both sales volume and sales value data for a longer period

The fourth condition, impact of price elasticity will be tested partly with available data, but first a theoretical assumption will be made.

3.2 Aftermarket

prices

As mentioned earlier in this research, it is not possible to prove aftermarket power in the aftermarket for laser printer via the theoretical manner of establishing that a company is charging a margin over its marginal costs for a number of reasons. To avoid this problem, this research attempts to prove market power via retail prices as they are set in these aftermarkets, by comparing price erosion in the aftermarket with the price erosion in the foremarket.

3.3

Limitations of the research model

the case of printer manufacturers, this trend has led to a number of technical restrictions in the design of toner cartridges. Printer vendors are not allowed to use techniques that support incompatibility, if it can be avoided. As a result, in general toner cartridge design is much more simplified. This makes it easier for third party manufacturers to copy the design and sell toner into the proprietary aftermarket of the printer manufacturer.

4.

Findings and conclusions

This chapter will discuss the testing of the conditions in the research model. It will delve into the market used for testing and the data will be explained and discussed. Next the four conditions will be tested. The chapter concludes with an evaluation if a company indeed has market power on its proprietary aftermarket.

4.1 The

market

The laser printer market is divided into many segments. Some of the characteristics that determine the various segments are widely accepted as industry standard. However, many other characteristics are up for discussion. This is mainly due to the fact that these characteristics themselves change over time. One example is list prices of printers. As discussed, the prices in foremarkets have been eroding over time. Therefore, if the market is segmented based on for example retail price, the objects in the segment will continuously change. A second segment characteristic which is debatable is which sales channels should be included in a certain segment. Different vendors use different sales channels to reach their target customers.

Additionally, for the purpose of this research a market is needed which is more or less stable for a numbers of years. The term stable assumes the opposite of an emerging market. The primary reason lies in the frequent fluctuation in emerging market circumstances which rapidly change over a short period of time. Moreover, it is difficult to establish if these changes are due to the rules of supply and demand, or other more specific causes.

Graph 4.1 below illustrates the size of the market for monochrome laser printer with a speed of max. 35 pages per minute in Germany (it is displayed per quarter in sold quantities) The data is indexed with base year: 1996 = 100. The graph clearly shows that the market experiences a seasonal influence. In order to examine this time series more closely, this seasonal component needs to be eliminated.

Graph 4.1 Market size per quarter

Mono market <35ppm - market size, per quarter

0 20 40 60 80 100 120 140 160 19 96 Q 1 Q2 Q3 Q4 19 97 Q 1 Q2 Q3 Q4 19 98 Q 1 Q2 Q3 Q4 19 99 Q 1 Q2 Q3 Q4 20 00 Q 1 Q2 Q3 Q4 20 01 Q 1 Q2 Q3 Q4 20 02 Q 1 Q2 Q3 Q4 20 03 Q 1 Q2 Q3 Q4 20 04 Q 1 Q2 Q3 Q4 20 05 Q 1 Q2 Q3 Q4 market size (1996 Q1=100)

Graph 4.2 below shows the same sales volume as shown in graph 5.1, however in graph 5.2 the sales are accumulated per year.

Graph 4.2 Market size per year

Mono market <35ppm - market size, per year

0 20 40 60 80 100 120 140 160 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 market size (1996=100)

This graph shows the actual trend in this segment of the market. In the IT market in general as well as in the printer market, cyclical trends are common and they usually follow the economic cycle of a country.

4.2

Testing the conditions

This chapter will discuss the conditions as earlier stated in the research model. If the four conditions hold, the test to see if a company has any market power can be assumed to be valid.

4.2.1 Foremarket

competition

The first condition is that the foremarket in which the company is active, is a competitive environment. This condition was set to ensure that the company has an incentive to actively search for other markets to increase its financial performance.

A simple but effective method to measure the market structure from a supply side is the Herfindahl index or H-index [Martin 2001]. It is calculated by accumulating the squared market shares of all companies in the market:

2 2 2 2 1 s ... sn s H = + + + also noted as

∑

Where si is the market share of company i and n the number of companies in the market. The

Graph 4.3 H-index and market size

Mono market <35ppm - market size and H-index, per year

0 20 40 60 80 100 120 140 160 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 ma rk et s iz e 0.0 0.1 0.2 0.3 0.4 0.5 H -index market size (1996=100) H-index

The graph shows an H-index decreasing from close to 0.4 to well below 0.25. This suggests increased competition, especially from 2001 onwards the market can be characterised as very competitive. Market share on the other hand is not increasing, and in ’02 and’03 even decreasing. The companies in the market, therefore, have to compete for a diminishing demand. A sanity check that can be utilised to test if the above theorem is valid, consists of taking a look at prices in the market.

Graph 4.4 on the next page shows the average price of the products as sold in the market. The average sales price in a year is calculated by dividing the total market value with the total quantity of products sold. Prices are adjusted for inflation17. Average sales price is displayed as an index with base year: 1996 = 100. As expected the average retail price decreases rapidly indicating a competitive environment.

17

Graph 4.4 Average retail price

Average retail price mono<35ppm - market size, per year

0 20 40 60 80 100 120 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Printer ARP (1996=100)

From the above it can be concluded that the foremarket for laser printers is indeed a competitive market. Both the H-index and the price trend in the market show that a company is clearly not in the position to exercise any kind of market power.

4.2.2 Link fore market vs. aftermarket

The link between foremarket and aftermarket cannot be tested with an existing dataset, such as a market structure. The reason for this is that there is no empirical research available for this market. Therefore, this condition has to be supported by theory and untested market experience. It can be assumed that there is no or a very weak link between fore and aftermarket. As also discussed in chapter two, this weak link is probably due to the limitations of the customer being unable to have a well balanced picture of all variables concerning its purchase. The customer cannot or does not want to consider aftermarket circumstances like for example toner prices when purchasing the foremarket product (the laser printer). Usually customers do not take the effort to investigate aftermarket products. This behaviour is stimulated by the vendor which is emphasizing foremarket product characteristics compared to aftermarket product characteristics in its marketing activities18.

For the sake of this research, it is assumed that there is no link between fore and aftermarket.

4.2.3 Third party supply in the aftermarket

The third condition in the research model is that there cannot be any competition that has considerable market power in the proprietary aftermarket of a vendor. If there are any serious competitors in the market, this limits the company’s ability to freely set prices.

Before going into the competition itself, first the total market for toners is displayed (see graph 4.5 on the next page). The dataset is indexed with base year: 1996 = 100. This is the market for toner, which can be used in the printers that are sold in the foremarket. Due to the incompatibility between aftermarket toners and foremarket printers between different brands, every vendor has his own proprietary aftermarket. What graph 4.5 actually shows is the accumulation of all proprietary after markets.

18

Graph 4.5 Market size for toner per year

Toner market mono<35ppm - market size, per year

0 20 40 60 80 100 120 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 market size (1996=100)

In order to test the market for third party toner manufacturers for every individual foremarket vendor, every individual proprietary aftermarket needs to tested. Since there are over 40 vendors active in the foremarket and therefore there are over 40 proprietary after markets it seems somewhat tedious and, more importantly, not significant to this research to test every proprietary aftermarket. This research focuses on two aftermarkets, firstly a large player, and secondly small to medium player. The size of the company refers to their respective market share in the foremarket for printers.

Graph 4.6 Market size and H-index for large player

Toner market mono<35ppm - market large player and H-index, per year

0 20 40 60 80 100 120 140 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 ma rk et s iz e 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 H-in de x market size (1996=100) H-index

Graph 4.7 Market size and H-index for small player.

Toner market mono<35ppm - market large player and H-index, per year

There is a clear trend recognisable in both markets. In both markets the H-index well above 0.7. The small player aftermarket even has an H-index that is well above 0.8 for most of the time in the time series. The H-index of the large company is a little lower that the H-index of the small company further suggesting that there is more competition in the large company aftermarket. The increased competition in the large company aftermarket refers back to the foremarket. As the large company sells more printers and has a larger installed base than the small company, its aftermarket is also larger than the aftermarket of the small company, making the large company’s aftermarket more interesting for third party manufacturers to enter.

The high H-index in both proprietary aftermarkets suggests that there is very limited competition in these aftermarkets. This means that there are no competitors in the market, which can limit the vendor’s ability to freely set prices. The condition is therefore met.

4.2.5 Price

elasticity

The last condition concerns price elasticity. The condition states that if prices in a market are highly elastic, the company has less flexibility to set market prices. As discussed in chapter two, the aftermarket for toner is different from other aftermarkets, because toner is needed to keep the system of goods functional where other aftermarkets deal with products that only improve the system. In the latter case, the aftermarket product is not necessarily needed to keep using the system of goods.

are adjusted for inflation19 and both time series are indexed with base: printer average price in year: 1996 = 100.

Graph 4.8 Aftermarket price vs. fore market price

Average price toner vs. average price printer

0 20 40 60 80 100 120 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Printer ARP (1996=100)

Toner ARP (printer 1996 = 100)

The chart shows that even after ten years, the average price level of printers in the market is still 3.5 times higher than the price level of toners.

Consequently, the consumer experiences (high) switching costs and as a result, price elasticity is very low. Consumers are not expected to switch to another brand under these circumstances. For the purpose of this research, it is assumed that consumers do not take aftermarket conditions into account when purchasing a foremarket product. Therefore, pricing of aftermarket products will not have an impact on the foremarket leaving the vendor the freedom to set prices in the aftermarket.

Note that the prices of toner seem to be very stable in this market. This is because the price in the graph represents the average price during a year. On a weekly basis, price toner prices in general

19

4.3 Aftermarket

power

Since no data is available to prove market power via a mark-up over marginal costs, a research model with four conditions has been created. In the previous paragraph established that these four conditions are met. Therefore, it can be tested if the vendor is able to set prices in its proprietary aftermarket.

The two graphs below demonstrate the following: the first graph (graph 4.9) shows the market prices in fore and aftermarket same as the graph in chapter 4.2 and The second graph (graph 4.10) shows the price erosion year by year in both the printer market and the market for toner.

Graph 4.9 Aftermarket price vs. fore market price

Average price toner vs. average price printer

0 20 40 60 80 100 120 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Printer ARP (1996=100)

Graph 4.10 Price erosion aftermarket vs. price erosion fore market Price erosion printer vs. toner

-20% -15% -10% -5% 0% 5% 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Printer ARP (1996=100)

Toner ARP (printer 1996 = 100)

In almost all IT markets in Europe and certainly in Germany, prices are subject to erosion. The graphs show that the toner market is clearly not subject to erosion. Prices in the aftermarkets are subject to change. In the long run, however, price changes stay within a bandwidth of five percent. It can, therefore, be concluded that the vendor has no pressure to lower its prices, while it experiences price pressure in its foremarket.

5.

Conclusion and further research

The research model set out to establish if a company that is experiencing competition in its main market, is ultimately able to create a profitable proprietary aftermarket. In the aftermarket, the company can then earn a margin on its sales where the competitive foremarket prevents it from doing so. This chapter summarises the outcome of the research, presents potential management decisions and provides some ideas for further research into this topic.

5.1

Evaluating the research model

This chapter evaluates the four conditions in the research model as well as the retail price test in the research model.

The first condition stated that companies in the printer market had to experience a competitive

foremarket. As conditions of a competitive market restricts the ability for a company to earn a

margin in its foremarket, it has to revise its business model and create other markets where it does earn a margin. The research shows that the market for Laser Beam Printers is very competitive and it can, therefore, be concluded that companies do have an incentive to create other markets to increase there financial performance.

If a company sets up an aftermarket, it has to meet three conditions in order for the company to be able to earn a margin over marginal costs. Firstly, there cannot be link between the aftermarket

and the foremarket. If the customer anticipates aftermarket conditions when purchasing a

foremarket product, the company is limited in its ability to freely set prices in the aftermarket. The research is not able to test this condition via empirical methods, because such data is not available for this market. Theory and market consensus, however, suggest that such a link does not exist.

plunging prices and the inability to earn margin. With reference to the available dataset, it is established that companies in fact have vary little competition in their proprietary aftermarkets. Thirdly, price elasticity in the aftermarket has to be relatively low. If the price elasticity is low, it provides the company more freedom to set prices in a market without experiencing a large change in the quantity demanded by the customers. In this particular proprietary aftermarket, customers have to buy the product for the system to stay functional. This means that if they want to switch to another brand’s aftermarket product, they will ultimately encounter high switching costs. The research established that as long as switching costs are higher than the costs to continue using the current brand’s product, price elasticity is low.

Since the four conditions are met, the research can establish if the company is in fact able to freely set prices in the aftermarket. This research strives to establish this by testing price erosion in the aftermarket versus price erosion in the foremarket. Almost all IT markets experience high price erosion. If a printer vendor has the ability to freely set prices in its proprietary aftermarket, this market cannot have the same amount of price erosion as in the foremarket. This research demonstrates that the aftermarket does not contain any price erosion and therefore, a printer vendor is flexible in setting prices for the aftermarket.

The four conditions were set to establish that setting prices in the aftermarket does not cause a negative market response. Since the conditions are all met, this is not the case. Therefore, a printer manufacturer can freely set prices in its proprietary aftermarket without causing a negative market response in its foremarket or its proprietary aftermarket.

5.2

Reviewing management dilemma and management question

can set prices, end therefore has significant market power, without causing negative market responses.

However, as mention in chapter one, the printer vendor faces continuously changing markets. Especially market dynamics in the aftermarkets are subject to change since those markets are still very profitable and therefore trigger entry. Since entry and especially blocking of entrants by incumbent firm is a topic on any government’s agenda, it requires a lot of creativity within companies to be able to react to changing market circumstances. It is therefore key for the printer vendor to continuously monitor and update it’s business model if needed.

5.3 Further

research

Throughout this research, a number of limitations where encountered. Additionally, a number of elements were highlighted, which are closely related, but were outside the scope of this research. The most important limitation of this research (other than availability of data) is that the research model did not include governmental influences. As stated before, national and European governments are acutely aware that a market which is not efficiently organised can be detrimental to their economy. As this research concludes that printer manufacturers in fact do have certain market power in their aftermarkets, it is only a matter of time before this subject is properly addressed by the government.

The link between aftermarket and foremarket is one that could not be empirically tested within the scope of this research. This is, however, a topic that can prove to be very interesting for further research, not only for printer manufacturers but also for many other companies which intend to set up aftermarkets. Firstly, as customers do not anticipate aftermarket circumstances when purchasing a foremarket product, it can be very difficult to establish a profitable aftermarket. Secondly, the ability of the consumer to see beyond the initial impact of its foremarket purchase might also strongly influence foremarket circumstances.