Make your own free website on

The European Automotive Industry

Industry Features

In this chapter we will describe the basic features of the industry, and discuss how they affect the level of competition and conduct of market actors.

As mentioned in section 2.4 we will use a modified version of Porters five forces to describe the basic features of the industry. The model is modified in order to concentrate on the factors that have special relevance for the automotive industry. Thus we will not discuss all the factors that are normally synonymous with Porters five forces, but rather select the factors that have special relevance and incorporate them into Porters framework.

Figure 3.1


It should be mentioned that this model has little analytical value in itself, and is merely a tool for structuring the presentation of the industry. Thus, in some cases we shall need other theories to examine the individual features. In these cases we will state the theories used, in the relevant section. For an overview of theories used, see section 2.4 on methodology.

As it is shown in the model we will not discuss substitutes as intended in Porters original model. The reason for this is that the substitutes available for private motoring have little effect on the nature and level of competition in the car industry. However if large monopoly power were to emerge in the industry, public transportation as a substitute may set limits on the prices of a monopoly producer. However, substitutes are needed in order to define the industry. Hence we have changed substitutes to industry definition. Of course this will be the first issue to be discussed below.

The features relating to barriers of entry will not only be discussed from the perspective of keeping out potential entrants, but also from the perspective of how these factors affect actors that are already in the industry. Under buyers we shall mainly focus on demand, both with respect to current quantity and with respect to the factors that influence the size of demand. This section will contain a large amount of empirical data regarding the size of the industry. Further we will discuss market power and vertical integration on the supply side as well as on the demand side. For the section of rivalry we will measure and examine the effects of concentration. Also we shall discuss overcapacity in the industry, as well as exit barriers and fixed costs.

Below industry definition, rivalry, buyers, suppliers and barriers to entry are discussed in that order. This order is not coincidental. There is an amount of interrelatedness between the individual factors of the model, and the order has been chosen in an attempt to describe the most basic structures first, and their dependants later. However, in some cases it may be necessary to refer to a later sections.

Below we shall briefly define the industry in question. In order to do so, we need to discuss not only the industry itself, but also the possible substitutes. An industry is merely a group of companies that produce a product with similar distinctive features. Thus we need to identify these features.

As mentioned in section 2.5 on limitation, this report focuses on the passenger car industry in Europe. The benefits provided to consumers of the industry products are easy transportation, and thus possible substitutes are other forms of transportation. In Western Europe the main substitute for cars is public transportation and non-motorised transportation like bicycles. Public transportation functions as a substitute for longer distances, whereas non-motorised transportation is a substitute on shorter distances.

The effect of the presence of substitutes depends on the closeness of these substitutes . That is how much they resemble the product in question. One of the main features for personal motorised vehicles relative to public transportation is convenience and flexibility. However these differences are decreasing, as public transportation is experiencing considerable growth as an effect of pollution pressures toward collective, low-emission forms of transportation. This has induced policy makers to set up more advanced schemes of public transportation, which allows for increased flexibility through more departures. Thus public transportation is an increasingly significant substitute to private car ownership at lower prices. This induces car manufacturers to produce cars that are more economic with respect to variable costs like fuel.

From the discussion above we have limited our focus to personal motorised transportation. However, even in this category there is a large amount of different companies, like component producers, car dealerships etc. This report will focus on companies that assemble the finished vehicle. All of these companies are also in the business of making engines, and designing vehicles. However this definition excludes companies that only produce car components, whereas assembling companies that are vertically integrated into component production are included.

Below we shall discuss some factors that affect the rivalry, and thus the fierceness of competition in the car industry.

Economic theory suggests that the vigour of competition is related positively to the number of firms in the relevant industry when other conditions such as height of barriers are equal. There are a number of reasons for this phenomenon.

One is that in highly concentrated industries, firms have a larger degree of market power. This is most obvious in the case of maximum concentration namely monopoly, but market power also occurs where there is more than one actor in a concentrated industry. It has been estimated that in an industry where the 4 largest firms control more than 40% of the market, oligopolistic competition is likely to occur .

The effect of such market power is that firms face a less elastic demand curve. Thus in order to sell one extra unit, the firm needs to reduce the price on all units, placing marginal revenue below the demand curve. As the firm produces at MC equals MR, price is higher than average costs. Thus there are monopoly rents for the firm. This is shown below in figure 3.2.

In a situation where industry concentration is at a minimum, substitution between firms will lead to a situation, where the individual firm faces highly elastic demand. This happens because firms cannot sell at a price significantly above the price of its competitors, and consequently competition will push prices towards the level of average costs. Accordingly economic rents are minimised.

Indsæt figure 3.2 fra side 19


Another reason that concentration increases monopoly power is that collusive behaviour is more likely to occur in highly concentrated industries . Hence in concentrated industries firms are more likely to agree on terms that limit the competition in the industry, thereby allowing firms to retrieve economic rents.

The question is how to measure the concentration in an industry. The optimal method would be to calculate the difference between marginal costs and price, as it is done in the Lerner index. Such a method directly measures the effect of concentration, rather than the concentration itself. However since we do not have exact information on marginal costs, we are forced to use alternative measures.

One of these is the CR4, which is the total market share of the 4 largest competitors in the market. In appendix A we have listed the different European competitors, and from this we can calculate CR4 to be equal to 52%. The problem with such a concentration ratio is that it only takes the 4 largest firms into account, and doesn't evaluate whether these are equally large, nor the size of competitors outside the top 4.

Therefore the Herfindahl-Hirschman index is a better measure, as it incorporates all market actors into the calculation. This is done by summing the squares of all the actors individual market shares . Thus in a monopoly HHI = 10.000, whereas an indefinite amount of firms lead to an HHI of 1.

For the European car industry HHI equals 1012 (Calculated on the basis of appendix A). This places the car industry somewhere in the middle with respect to concentration. Thus there is a basis for some market power and limited competition. However HHI is not a measure of monopoly power like the Lerner index, and therefore the existence of a possibility for market power, does not necessarily mean that such power exists.

It is interesting that in the U.S, the concentration of the industry for passenger cars is much larger. Here CR4 equals 97 percent .

With a CR4 of 52 % and an HHI of 1012 oligopolistic competition is likely to emerge as the industry market structure. This indicates that there is a small amount of market power, and that the individual firm is affected by pricing decisions of the other companies. However, an HHI of 1012 is still small enough to ensure competition. The reason for this relatively low HHI is that the dominant players in the industry are of approximately the same size.

A major problem for the European car industry is overcapacity. Even if the terms "production capacity" or "potential production capacity" are not easy to define or to quantify, these terms are very important to understand when discussing the terms regarding excess production capacity. In some cases the overall design capacity of a plant is higher than what we would call the maximum technical capacity. However, the production capacity and related terms has been defined as described below :

> The ideal or maximum capacity is the level of output that would be attained if circumstances were ideal. Sometimes also known as full capacity.
> The practical capacity is the level at which, it is reasonable to expect the factory to operate given the time lost for repairs to equipment, holidays and so on.
> The normal capacity is the average level of production needed just to fulfil existing demand for the company's output.
> Excess capacity is the amount of extra volume that a firm could produce, if all it's existing plant and equipment were fully used for 24 hours a day. Hence excess capacity equals practical capacity minus normal capacity.

If all Europe's plants were manned and equipped to run at full stretch, the excess capacity would be 3.5 million cars a year . To make a bad situation worse for today's manufacturers, Japanese factories in Britain are building up towards a production of one million cars a year by the end of 1990's. Demand for the passenger cars are considerable lower than earlier anticipated. Excess capacity will therefore not be cancelled by increasing demand. Hence prices will come under even more pressure. This in turn will raise break- even levels and aggravate the effects of excess capacity. In 1995 Western European carmakers produced 12.9 million cars from their total capacity of 18.1 million. This represents a low capacity utilisation of 71.2 per cent.

In a recent speech, George Simpson, former Chairman of Rover and now Chairman of Lucas, estimates overcapacity to be as high as one third and is setting the scene for restructuring and mergers. The combination of excess production capacity and lower than expected demand, however, clearly shows that structural changes are inevitable in order to balance demand and supply in the near future.

In spite of this existing excess the European manufacturers are talking about new investments to increase their capacity. Accordingly they plan to expand their capacity by close to one million units by year 2000. Total capacity, including further increase of capacity at existing Japanese, Eastern European and Asian transplants, is expected to be 19.1 million units by the year 2000. This will be an increase of 5.5% compared to the capacity in 1995, which were 18.1 million units .

It can be discussed why the problem of overcapacity has arisen, and why further investment occurs in spite of this. It appears irrational, that many firms wish to increase capacity in an industry that already suffers from excess capacity. One possible reason is that the management of the individual manufacturers realise that the industry is declining, and that there is a need for some companies to exit the industry. Therefore everyone invests in order to ensure that they are one of the companies that survive the industry decline. Michael Jensen supports this theory in his article "The Modern Industrial revolution, Exit, and the Failure of Internal Control Systems", where he argues that in industries with excess capacity, where managers should be concerned with downsizing, they instead continue to invest leaving downsizing to other industry actors.

Another reason for overinvestment is that it may be a consequence of game theory, where each of the market actors seek to demonstrate their commitment to the industry by investing heavily, thus deterring others from making investments. This leads to an environment, where all the market actors hurry to invest in plants or other sunk costs, even though it appears to be irrational in light of the existing overcapacity. An example of how such game theory might work in the car industry is given in the box below.

Indsæt billede fra s. 23


In all circumstances the existence of overcapacity is a powerful catalyst for fierce competition. In such an environment the marginal cost of producing an extra unit is relatively low, and the break-even level in terms of units sold is relatively high. Thus the firms in the industry have a strong incentive to cut prices, in order to stimulate demand, and consequently putting their excess capacity to use.

Exit barriers and value of fixed costs
Another determinant of competition in the car industry is exit barriers and the value of fixed costs. These are somewhat interrelated since fixed costs relating to specialised assets is an exit barrier.

The effect of high exit barriers is that firms tend to stay in the industry, even if it is not profitable. This leads to increased competition, as some market actors are willing to operate at a loss in order to avoid the high costs of exit. Such exit barriers can stem from a number of factors, like for example contracts that are expensive to break or specialised assets as mentioned above.

In the car industry there are a number of highly specialised assets like production plants and the value of knowledge in the form of R&D (R&D is quantified in chapter 7). Such assets cannot be transferred to other industries without loosing value. Thus if a firm chooses to leave the industry, it will loose the value of its specialised assets. This is not the same as to say that existing firms should always stay in the car industry. If the firm is running at a loss, and there is no chance of turning this situation, it will be better to incur the cost of loosing its assets right away. However if the deficit is perceived to be temporary, the value of assets will incite the company to stay in the industry, in the expectation of future profits.

In the car industry, where there are a number of large producers (see appendix A), industry specific assets are likely to be traded among industry actors at a reasonable price. Thus there doesn't seem to be much of a case for the high exit barriers argument in the car industry.

High fixed costs are another catalyst for competition. High fixed costs often lead to low variable costs and marginal costs. Thus firms are induced to cut costs, in order to sell more, so as to cover their high proportion of fixed costs. Fixed costs will be further examined and quantified in chapter 7. Here we show, that the amount of capital to labour is relatively high in the industry, but at the same time labour costs, far exceed depreciations. Thus although fixed costs are high there appear to be no extreme threat for price wars. Both fixed costs and exit barriers are related to the game theory example above, since sunk costs are necessary to make a credible commitment to the industry.

Industry growth
In a situation where the size of an industry is declining, there is a struggle, to be an industry survivor. Hence low growth leads to intense competition, whereas high growth diverts the attention of companies towards capturing the shares of the new market, so that competition is less head-on.

After almost fifty years of steady post-war growth, only interrupted by the oil crises in 1973 and the early eighties, car demand suddenly dropped by a massive 16.8% in 1993. Only 11.3 million cars were sold in 1993. This is nearly 2.1 million units less than the previous year. In other words a loss of 2,5 billion ECU, which is equivalent to the total output of all Western European Volkswagen plants. The development is shown in figure 3.3 below.

Indsæt figur 3.3 fra side 25

With the benefit of hindsight, the 1993 trend line break should not have been as unpredictable as it looked at the time. In reality demand had started to loose its edge already a few years earlier. However at that time the change in growth was concealed by two extraordinary events that temporarily boosted demand.

In 1991, after the reunification of Germany, West- European registrations enjoyed an almost overnight windfall of an additional 700,000 East-German registrations. Together with the beneficial effects on West-German new car sales, triggered off by a massive export of second hand cars to former East- Germany.

Secondly, during the years that preceded their EC membership and during the first years of membership Southern European market like Spain, Greece and Portugal experienced a strong increase in passenger car sales. During the period 1984- 1989 the Spanish car market tripled from 500.000 in 1994 to 1.5 million in 1989. Demand eased down again to previous levels after the average age of car parc had settled at a lower level, more in line with the newly found wealth and increases in GDP per head.

As it is shown in section 3.3.2 demand is correlated to economic prosperity. In the table below we have shown some economic indicators for EU15.

Table 3.1
Source: Neil Mulliniuex: The new car market in Europe & OECD Economic Outlook.
Figures after 1996 are projected.

GDP per head has been increasing by a considerable percentage. However, GDP per head figures can be misleading because these figures do not consider differences in the cost of living, so it does not show the real standard of living. We therefore use the real GDP growth to measure economic growth, which includes differences in the cost of living. In 1994, for example, real GDP growth was only 2,6%, but growth of GDP per head was 6,4%. Real growth of GDP continued to improve in 1994 and 1995 after a period of low growth between 1991 and 1993. The economy of the region was estimated to have grown by 2,6% - 2,7% in each year. Though the European economy had below average rate of world trade growth, it was a good performance from such mature economies.

Table 3.2
Source: Molle, Willem: The Economics of European Integration; page 439

PPP = Purchasing Power Parity

If we look at the European economy, the average for the countries that now make up the EU15 was almost three per cent a year over the last forty years. First there was a long period of high and stable growth (four per cent a year) that lasted up to the first oil crisis. This coincided with the first extension of EU. Next there was a period of adjustments, with low growth (1,5 percent a year). The end of this period coincided with the second extension. Finally the period since 1985 is one of new dynamism (growth rate of 2.5 per cent a year).

If we need to account for the differences in the cost of living, we can use Purchasing Power Parity (PPP) index. This index allows for more direct comparison of living standards in different countries. In the above table PPP is set equal to 100 for the average purchasing power in EU member countries. According to table 3.8, Luxembourg had the highest standard of living, PPP 165, in1995, while Greece had the lowest standard of living, PPP 63, in EU. It means that an average citizen of Luxembourg's purchasing power was nearly three times higher than an average citizen of Greece.

All in all projected growth in Europe is moderate. Certainly the potential for growth is not as big, as it tends to be projected in the annual reports of the manufacturers. Combined with saturated markets it does not appear that demand for cars is likely to increase significantly in Europe. This low growth environment is likely to increase competition in the industry. However there may be a possibility for high growth in the East European countries, if some of these are accepted for an EU membership. Acceptance may lead to rapid Economic growth in these areas, and hence a temporary boost in car demand, like the one in Southern Europe in the late 80's.

Below we shall discuss the elements in the buyer section. These include vertical integration, market power and demand. Demand has been subdivided into three sections. The first, structure of demand, presents empirical data on the sales of automotive vehicles in different countries. The second section discusses the factors that drive demand, while the third section discusses the effect of segmented markets.

Structure of Demand
The table below shows the unit sales of passenger cars in 17 European countries in the period from 1991 to 1996.

Figure 3.3
Source: Neil Mullinieux: The new car market in Europe; page 5

According to the table, the European total market grew by 6,3% in 1996 and the growing convergence of the European economies resulted in all countries showing positive growth over the previous year. However, it is clear that all the markets are mature to a greater or lesser degree. Only the Norwegian car market is growing strongly. In other words, it seems to be recovering to original levels from a deep depression. There is still some overall growth potential in the southern markets of Portugal, Spain and Greece, but even here the sale of passenger cars is influenced far more by the immediate health of the economy than by longer-term growth in GDP.

The five largest countries, namely Germany, France, Italy, UK and Spain, dominate the European demand for automotive vehicles, which is demonstrated in figure 3.4 below. Together these countries accounted for 80,4% of car sales in 1996.

Indsæt figure 3.4 fra side 30


Despite a fall in sales from 4,158,674 units in 1991 to 3,496,320 units in 1996, Germany remained in a dominant position in the market, accounting for 27.4% of all sales. The German car market was already the largest in Europe before unification and it has now risen to a new level. It is the key market for any manufacturer, not just because of its size, but because it consists of larger and more expensive vehicles than most other countries .

France, Italy and the UK are markets of almost the same size, with annual demand of nearly 2 million vehicles in each market, each having approximately 15% of the European total market share. While the UK, Spanish and French car markets recovered in 1994, the demand for new cars declined further in Germany and Italy. Recovery in French and Spanish market was boosted by government-sponsored incentives for car replacement. Over the last two years several governments have adopted the French idea of offering incentives to stimulate the sales of passenger cars. Usually this has taken the form of direct reduction in the price of a new car, if cars of certain age were being replaced. The stated rationale for this type of incentive was usually environmental .

In Western Europe car ownership is about 20% below the level in the USA, although the key markets of Germany, Italy and France are much nearer to the American average . There are some other factors, which will prevent car ownership in Europe ever reaching the level in the USA. Europe has a smaller land area and a larger population compared to USA. The average population density is higher and this inhibits car ownership, particularly as more people live in cities. By their very nature cities discourage the use of cars by making it difficult to park, while they have enough public transport services. Further more, most European cities are much older than cities in the USA. Therefore it is not easy for European cities to fulfil the special requirements of wide and straight roads and plenty of parking space.

The European passenger car market, however, is large in production and in turn over. It contributes nearly 1/3 of the global car market. Worldwide sales of new cars increased from 33,1 million units in 1993 to about 35,0 million in 1994. At the same time in the EU market, sales recovered to 11,2 million units in 1994 from 10,7 million units in 1993. In 1996 sales of new car in the European market increased to 12,8 million from 12,0 million in 1995. Even though this 6,3% of growth in the market was extraordinary, the manufacturers did not seem to be well and many of them experienced considerable problems in terms of low margins and high marketing expenses . This is an indication of tough competition.

Demand Factors
In the following pages we will aim to identify the factors drive demand, and examine them. Where empirical data on a suggested factor is available, we shall also seek to demonstrate the relationship to demand.

Table 3.4
Source: Neil Mullinieux: The New Car Market in Europe; Page 230

In order to do so, it is necessary to begin with a few definitions relating to the size of demand. The vehicle parc is the number of vehicles in use within a market. Approximately 75 percent of the world's vehicle parc are within developed markets, and it is set to exceed one billion vehicles in 2015, of which just 59 percent will be in developed markets. Vehicle parc change slowly, as parc dynamics revolve around a current average vehicle life of 6-7 years and overall parc renewal every 13 to 14 years. The rate of growth in vehicle parc is the key determinant of the future structure of the world's auto industry. This growth and associated volume are determined by sales and scrappage in individual countries, depending on level of development in each country. North America and Western Europe have been dominant players as measured by their combined share of the world GDP. Vehicle density is the number of vehicles in use per person in a country. In 1995 the average vehicle density in the world was 117 vehicles per 1000 of the world population. In the developed world, density was 547 per 1000 people, while in the developing world it was estimated as 34 per 1000. Thus vehicle parc equals density times population. Table 3.4 above shows data on these terms for sixteen European countries.

Thus the primary factors that drive the demand are population size, vehicle density and average vehicle life. We expect vehicle life to depend on GDP pr. capita and the price of cars in the country in question, since wealthier populations tend to drive larger cars. However we lack the empirical data to test this.

Vehicle density is determined by a number of factors. Undoubtedly one of these is the price of vehicles. This price is often distorted by a number of taxes on vehicles and on the use of vehicles. Due to the variety of government imposed expenses on vehicle use, it is not easy to get a clear picture of the tax pressure in different countries.

Further there is reason to believe, that vehicle density is related to GDP pr. capita, and to population density. These relationships are examined in the two figures below. The figures are made of data from table 3.4 and appendix B, by plotting GDP pr capita and population density, against vehicle density.

Cars are regarded as a luxury good, and thus it is likely that the vehicle density is highly affected by the wealth of a country's population. This relationship is demonstrated in the figure below. Here there seems to be a strong positive relationship, between wealth and vehicle density.

Indsæt Figure 3.5 fra s. 34

Another hypothesis is that the level of car ownership is related to the population density of a country. The reason for this is, that in areas with high population density there is less need for long distance transportation, and public transportation is often more advanced in these areas. However as it appears from the figure below there seems to be no empirical evidence of this hypothesis. Quite surprisingly there seems to be a positive relationship between population- and vehicle density.


Indsæt Figure 3.6 fra s. 34


However, as it can be seen there is a very large variation in the data. This implies that the tendency towards a positive relationship between population- and vehicle density is unlikely to be statistically significant. Thus we cannot conclude that there is any relationship between the two in Europe.

For both of the graphs there is the problem, that the data is influenced by other factors, than the one measured (e.g. taxes). This is especially a problem in this case, where we only have 16 observations, which is not enough to ensure that factors are not biased towards a specific result.

Apart from the factors mentioned above, the demand for cars, like all other products, is affected by substitutes and complementary goods. As mentioned in section 3.1 the primary substitute for passenger cars is public transportation. Consequently if public transportation becomes more competitive through better service or lower prices, demand for cars will be lowered.

Likewise if the prices of complementary goods are lowered, demand for cars will increase. Among the most important complementary goods are fuel and road taxes (In some countries). A less obvious complementary good is finance. Cars are a major expenditure in the budgets of most people, and thus they often need some sort of external finance in order to purchase cars. Hence the interest rate will affect the demand for cars. This relationship has caused the largest car manufacturers to integrate into financing, thus allowing more control of this factor. This is discussed in section 3.3.4.

The effect of segmented markets
Below we shall discuss the causes, evidence and effects of segmentation.

Under an agreement which came into force on January 1, 1993 the European Union (EU) and the European Free Trade Association (EFTA) have effectively merged to create the European Economic Area (EEA), which extends many of the provisions of the post- 1992 single market to EFTA member states . This has been labelled the world's largest integrated market.

However, the market is still highly segmented into national markets. This segmentation is caused by a number of factors. While economic integration may have created a market of 380 million consumers, those consumers have very different levels of purchasing power. The disparities are demonstrated in table 3.2. One of the central aims of the EU and the wider EEA is to reduce disparities between regions through structural aid. The main targets for structural aid are Ireland, Portugal, Spain and Greece. On top of these economic inequalities there are significant differences in climate topography, roads infrastructure, population density and culture, which combine to market diversity.

No matter what the reason, the fact is that the European market at present is very segmented, even though there is a trend towards increased integration of the European market.

Table 3.5
Source: European Commission; Car prices within the European Union on 1 November 1995.
Notification: B=Belgium; D=Denmark; E=Spain; F=France; IRL=Ireland; I=Italy; L=Luxembourg NL=Nethrlands; A=Austria; P=Portugal; S=Sweden; UK=United Kingdom.

Evidence of this segmentation can be found in the prices of cars across Europe. The table above shows prices of seven different car models across Europe. Prices are calculated free of tax as an index relative to the cheapest country. The figures are from November 1, 1995. The brands selected for the table has been chosen, in order to ensure maximum diversity across manufacturers and segments. Thus we have chosen seven models from seven manufacturers across six segments.

The table clearly shows that there are large differences in prices across Europe. Further it shows that there are no clear patterns with respect to cheap and expensive countries. In an integrated market price differences should not exist, as higher prices in one country would encourage the population to buy their cars in other countries, thus levelling prices throughout Europe. Since this is not happening, it must be because there are barriers, and hence markets must be segmented.

The major effect of the segmented markets is that it allows firms to price discriminate. In section 3.2.1 we discussed how a firm facing a downwardsloping demand curve, will need to lower the price on all units in order to sell an additional unit. In a segmented market the firm can select a price in each market independent from other markets. Thus if a manufacturer wishes to sell an additional unit, he can do so by lowering the price of all units in just that market.

A manufacturer that is able to price discriminate perfectly sets a price on each car sold equal to the maximum the individual customer is willing to pay (first degree price discrimination), whereas a non-discriminating manufacturer will set the price at the level the marginal customer is willing to pay. This can be illustrated through the demand curve. The downwardsloping demand curve indicates, that the first customers are willing to pay more than the customers further to the right on the demand curve. Consequently in a situation where there is only one price, some customers are paying less, than what they would be willing to pay if they had to. However a perfectly discriminating manufacturer can extract this surplus for himself and thus earn higher profits. This is demonstrated in the figure below.

Indsæt Figure 3.7 fra side 38


Segmented markets allow price discrimination on the basis of national markets, and as it is shown in the figure this can be used to get higher profits. However as it is also shown, the benefit from discrimination depends on the slope of the demand curve or rather the elasticity of demand. In infinitely elastic markets consumers are willing to pay approximately the same amount for different products, and thus there is no basis for discrimination.

Since the slope of the demand curve faced by the individual manufacturer depends on the level of industry competition as argued in section 2.1, it depends on all the other issues addressed in this chapter. For example a manufacturer in a highly concentrated industry can more easily benefit from price discrimination than a manufacturer in a more differentiated industry, where competition sets limits for the prices the manufacturer can charge. In the section summary we argue, that competition in the industry is high, leading to rather elastic demand. This decreases the gains from price discrimination.

Obviously no manufacturers in the car industry are able to discriminate perfectly, that is on the background of the individual consumer's willingness to pay. The car manufacturers are merely able to discriminate on the basis of national markets, and thus the size of the dark shaded area in figure 3.7 becomes much smaller. The exact size depends on the shape of the demand curve in the individual markets. Thus due to price discrimination the manufacturers can set optimum prices in each market, which yields higher profits than optimising from an aggregate point of view. Consequently prices will be set higher in countries where competition is weak and consumers are wealthy with elastic demand.

The effect of segmented markets on competition can be discussed. It is obvious, that segmented markets limit the extent of price wars. If a price war occurs in one national market, this price war does not automatically transfer to all other markets. Thus the cost of price wars are lower in segmented market.

It can be argued that the lower cost of price wars increases the likelihood of such wars occurring, thus increasing competition. On the other hand segmented markets may also deter smaller competitors from challenging larger firms. For example if a small manufacturer enters the Dutch market, at a low price, the larger competitors can relatively cheaply cut prices in that market in order to outcompete the new entrant in a segmented Europe. However in an integrated market, the large firm would have to cut prices in all markets, thus making the move more expensive. Thus competition from smaller competitors are limited in segmented markets.

Market Power and Vertical Integration on the Demand Side
The buyers of the automotive industry are a highly dispersed group. Most of the cars produced end up in private ownership, even though a few are sold to rental or leasing companies. Also many companies provide cars for employees that need the vehicle in their daily work.

Most cars are sold through dealerships, which are in most cases not owned by a manufacturer, even though they often have strong relations to specific manufacturers . The investments made by these dealerships are quite small, since they are not required to do any R&D or investment in any specialised assets. Hence if dealers are dissatisfied with the manufacturer, or the other way around, they can quite easily swap to another. Thus there exist little buyer power in the relationship between the manufacturers and dealerships.

Due to this low market power, transactions are often done at arms-length, and there is little need for vertical integration into dealerships .

However in some ways the manufacturers have invested into distribution. Thus all of the seven largest manufacturers described in chapter 5 have established financing companies . These companies provide finance for the individual buyers, as well as the dealerships, which need to finance their stock of vehicles.

The financing part is very important in the decision whether to buy a car, and in most instances the cost of financing is part of the total cost of the car. Thus owning financing companies, allow the manufacturers more control over the cost of the cars. Also car manufacturers are usually, as mentioned in the introduction very large companies with deep pockets, and thus they are able to provide this finance.

Another important feature of the integration of financing companies is that it is an alternative to vertical integration into dealerships. Since the owners of these dealerships usually do not have the funds to invest in a stock of cars and parts some sort of financing is needed. The financing companies of the large manufacturers provide this finance. Without these vertical integration into dealerships might be necessary.

Further some of the companies have invested in car rental companies. An example is Ford that bought Hertz in 1987 . This action can hardly be explained from a market power approach, but rather that Ford wished to ensure a secure demand for their vehicles, while entering a profitable business.

It is not the intent of this report to conduct a thorough analysis of the car components industry. Instead we will merely point out the special relationship between component manufacturers and assembly companies.

For this we shall use the theory of transaction costs, as a determinant of vertical integration. We shall also point out the relationship between market power and vertical integration.

Further we shall briefly discuss the barriers to forward integration by component manufacturers, and the importance of good supplier relations to competitiveness.

Market Power and Vertical Integration
Market power exists when one side of a relationship is more dependent on the other, than the other way around. E.g. A is more dependent on B, than B is on A.

Market power often occurs as a consequence of transaction costs. In the car industry transaction costs occur due to asset specificity. Often component suppliers are required to invest in very specific assets, in order to produce the components for a specific manufacturer . This specificity of assets lead to the possibility of opportunistic behaviour by the manufacturers, as they face the opportunity of demanding lower prices after the specific investment has been made.

The reaction of component producers is to secure themselves against this risk. This can be done through charging a premium or demanding a long-term contract. In both cases there are transaction costs, as the long-term contract doesn't allow for flexibility in either price or product, as the industry changes over time.

Thus the required investment in specific assets lead to market power, since manufacturers may act opportunistically. This leads to transaction costs, as component producers seek to protect themselves from opportunistic behaviour. One way to eliminate the chance of such behaviour is vertical integration since it removes the incentive for opportunistic behaviour.

Consequently vertical integration in the car industry is much more likely on the supply side than on the demand side. Manufacturers are especially likely to integrate vertically into components that are specific for the single manufacturer, and require large investment in assets. In chapter 7 we have tried to quantify the level of vertical integration for 4 manufacturers.

An alternative to vertical integration, is close co-operation with component manufacturers, which is an integral part of the Japanese Just in time system. We will not discuss this phenomenon any further.

Barriers to Forward Integration
In the case where the component manufacturers have the opportunity to integrate forward into assembly of cars this would increase competition. However, all of the component producers face the same barriers as everyone else. These include economies of scale, R&D and branding.

Further if component producers attempt forward integration, they will do so at the risk of loosing existing revenue, as assembly manufacturers are unlikely to help a potential competitor.

The Importance of Supplier Relations
Supplier relations are very important to car manufacturers, as they are needed in the car aftermarket.

As a consequence of vertical integration, car manufacturers make some of their own components. The production of these components is under the control of the manufacturer. However production of components sourced from other firms may pose a problem in the aftermarket. The aftermarket is the market for components and service after the initial sale of the automobile. Problems may arise if these suppliers choose to discontinue production of parts that are used in vehicles no longer in production. Since these vehicles are still in use, there is still a small demand for such parts, in the case of repairs. Often this demand is too small to sustain production of these parts, however they are essential to the manufacturer, because the price and availability of aftermarket sales and services of older models affect the sale of new models . Thus car manufacturers are dependent on their relationship to component suppliers, with respect to the availability of parts for repair. This is a further reason for vertical integration.

Barriers to entry are highly important to the profitability and competition of an industry, since low barriers to entry will encourage new entrants in the industry, until economic profits no longer exist. Thus competition is lowered through high barriers to entry. Below we shall discuss the barriers that are relevant to the car industry.

Economies of Scale
Economies of scale are often cited as a factor leading to entry barriers. In industries where economies of scale are significant, firms need a large volume of production in order to stay in the industry. If a company does not have such volume, it will face higher average costs, than other firms, and will eventually perish under the forces of competition from larger market actors.

Economies of scale occur when companies face a downwardsloping average cost curve. However as it is indicated in figure 3.8 below, average cost usually only drops for a while, before it flattens and eventually starts to rise.

Figure 3.8

Economies of scale work as an entry barrier, because firms who wish to enter the industry, must have enough volume to produce at the flat part of the slope. The minimum required volume is called the minimum efficiency scale (MES).

One type of economies of scale arises as a consequence of fixed costs. Often if production is large enough, it is profitable to invest in a more efficient production plant, which lowers the marginal costs, but increases the fixed costs.

Scherer and Ross suggest that minimum efficiency scale for a car production plant is 200.000 units . Obviously this is a very rough estimate, since it doesn't take into account that different car models are likely to require different production plants. However, minimum efficiency scale is very difficult to estimate accurately, but 200.000 units are not very much taking into account, that Ford alone produced 7 million cars in 1997 worldwide . Further evidence against product economies of scale as an entry barrier can be found in the fact that most manufacturers operate more than one plant. Thus it appears that economies of scale is only an entry barrier for the smallest and often specialised companies. Hence economies of scale might be an argument for producing a wide range of vehicles in different segments, since many of the factory operations are the same, no matter which model is being made.

For larger companies there are no product economies of scale, since these manufacturers far exceed the minimum efficiency scale of 200.000 units. However there is still theoretical evidence that there are economies of scale for large manufacturers, which will be discussed in chapter 8. These economies of scale are likely to relate to fixed and sunk costs, which do not relate to quantity of output. Here we think especially of R&D, but also of marketing.

R&D will be discussed separately below, and quantified in chapter 7.

Research and Development
On average the five manufacturers Ford, GM and Volkswagen spent 5,8% of sales (weighted avg.) on R&D (See Section 7.4 for details). This is way above the average industry, which spends approximately 0,8% of sales on R&D . For comparison the average in ethical drugs is 10,2% and in semiconductors 6,1%.

Although an industry average for the car industry has not been available, it appears that the car industry is very intense on R&D. The reason for this can be found in the special industrial structure. Throughout this chapter we have discussed factors that limit and facilitate competition. In the section summary below it becomes clear that the car industry is basically highly competitive, but at the same time there are factors that contribute to a bit of market power for the individual firm.

This is a structure that is ideal for facilitating R&D spending. In a situation of perfect competition where the individual firm faces a perfectly elastic demand, there is no incentive for companies to spend money on R&D, since all firms will earn the same profits. However in such a situation a revolutionary innovation, might help to change the market structure. On the other hand in a situation, where the individual firms have large market power, and face little competition, there is little incentive for innovation and R&D . This thesis is based on a Schumpeterian argument, where disruption and threat is a catalyst of innovation. Such disruptions are small in established monopolies.

Another interesting aspect of the average percentage of R&D for the companies above is that it is above the industry average. Since these companies are among the largest in the industry, it seems that R&D is necessary for growth and success in the industry. This gives R&D a role as an entry barrier, since most small companies cannot afford the large expenditures connected with R&D. At the same time R&D costs are sunk. Thus there are significant economies of scale connected to R&D, which further facilitates its role as a barrier to entry.

Brand Identity
Brand identity is closely related to marketing expenditure. For Ford and General Motors this equals approximately 2,4% of sales revenue. In 1977 the average industry spend 0,66% of revenue on advertising , but this figure has probably risen as a consequence of the increased public attention towards mass media. Thus the car industry appears to have marketing expenditures close to the average industry.

However Brand identity may still work as a powerful deterrent to new entrants. The purchase of a car is a major expense for most potential customers. Therefore it is important for the customer to be certain, that he is buying a quality product, and that he can acquire spare parts and service should a problem arise. Consequently it is important for manufacturers to have a strong public profile of stability and quality. Well-established companies can provide this more easily than newcomers.

Due to the combination of high price and a wide variety of manufacturers and product lines, potential buyers are likely to involve themselves in the purchase. Hence it becomes important for the manufacturers to provide information about their products to their customers. This further increases the importance of marketing.

Product differences
R&D and marketing both contribute to product differences in the car industry. R&D because it allows the manufacturers to develop new and different models, marketing because it gives consumers different perceptions of the manufacturers in the industry.

Thus the internal substitutability in the automotive industry becomes imperfect. This means that consumers are not indifferent to which car they purchase. Further it means that the car market becomes less transparent, and thus it is difficult for the consumers to evaluate how to get most value for their money.

In a situation of imperfect substitutability each manufacturer faces a slightly downwardsloping demand curve, or rather the manufacturers have some power over prices. As mentioned in section 3.2.1 a downwardsloping demand curve indicates some market power and thus some monopoly rents. Consequently product development and marketing lead to product differences, which again lead to market power. This market power decreases the level of competition somewhat, and keeps the industry away from a situation of almost perfect competition.

Government Policy
Several important aspects of automotive design are subject to government regulatory standards. While regulatory standards are primarily aimed at achieving two common social objectives, vehicle safety and environmental protection, they differ greatly from one market to another. Nations and regions independently have developed their own automotive regulatory standards with accompanying certification and testing procedures over time. As the automobile marketplace has become increasingly global, differing regulatory standards have become a major barrier to trade. Thus these regulatory differences are highly important in segmenting the European market (See section 3.3.3).

Compliance with multiple regulatory frameworks reduces vehicle affordability as it imposes substantial cost penalties and design manufacturing constraints. It is fundamentally inconsistent with free trade in a global automobile market. It also extends the time needed to develop new products, preventing manufacturers from responding quickly to the changing needs of consumers' world-wide. As it will be argued in the next chapter, exactly the quick response is an important source of competitiveness.

Additional costs incurred to design and develop different versions of a particular model simply to meet different regulatory and certification requirements may add more than 10 percent to the design and development cost of the product. Higher costs mean higher prices for consumers and reduced choice of products

There are two mechanisms by which government intervention is contributing to the emerging shape of the automotive industry: environmental regulation and co-founded R&D programs.

With regard to environmental issues, certain town and city centers have been closed to traffic, and incentive schemes have been set up. Among these are cars fitted with catalytic converters, scrappage incentives, and electric vehicles .

With regard to R&D it is essential that public authorities support the research and development efforts of the industry. EU has a key role in stimulating joint research and technological level between companies within the industry as well as on a wider multisectoral basis. It is to Europe that credit must be given for the major advances of recent years (injection, ABS, power assisted steering automatic gearbox, converter, etc.).

Europe's technological capability in the automotive sector is great. But in the field of research and development going beyond basic research, European co-operation in global terms is lagging behind the United States and Japan .

State aids remain a controversial area of the EUs competition policy. The actual orientation of governments is for the reform of state aid control and for more transparency in the Commission's decision-making process for the approval of state aids granted by Community governments to public and privately owned enterprises.

State aid weakens the competitiveness of the European industry. First it can lead to distortions within the market, which lead to inappropriate levels of outputs and perhaps inefficiency. Secondly, it can distort the nature of the oligopoly game, leading to unfair advantages or if governments engage in matching each other in the aid given to their own national firms socially wasteful expenditure .

The reason for using state aid in the car industry, is that it employs a lot of people, contributes a lot to a country's Gross Domestic Product and car makers want to sell cars and will invest behind barriers to do so.

All in all government policy has large effects on competition. The different standards in the different countries help to segment the market, making price discrimination easier, while state aid works as an entry barrier to keep out new potential entrants

In this section we have used Porters framework to evaluate different features of the industry, with respect to their effect on the level of competition. Further we have established some relationships between these features and the conduct of market actors.

Substitutes were used to define the industry as the companies that assemble motorised vehicles for personal use. We further argued that the closest substitute is public transportation, but that this poses no real threat to the industry at present.

Further we have shown that the European car industry is characterised by a moderate concentration, which allows the manufacturers little market power. There is a large amount of overcapacity in the industry, which induces heavy competition. In spite of this manufacturers are still investing in production plants, which we suggest is a consequence of game theory and strategic thinking. Although the sunk costs are relatively high in the industry, this doesn't lead to high exit barriers, because production assets are industry rather than firm specific. Thus the companies are able to sell their assets at a reasonable price. The markets for cars in Western Europe are saturated, and thus industry growth is rather small, which leads to tough competition for the existing customers. The fixed costs of the industry are, however, an incentive to cut costs.

Germany dominates the demand in the car industry, followed by France, Italy, UK and Spain. The main factor driving demand is economic prosperity. The buyers of cars are highly dispersed, and thus have little power. The segmentation of the European market has allowed the use of price discrimination, which increases profits.

Due to specialised assets there are a larger potential for vertical integration on the supply side of the industry. Car manufacturers are somewhat dependent on their suppliers for components for vehicles that are no longer in production.

The main barrier to entry is the high costs of R&D, followed by the importance of a good reputation. Economies of scale in production do exist, but with a minimum efficiency scale of approximately 200.000 units, it only affects smaller corporations.

R&D and brand identity both contribute to product differences, which allows the manufacturers a little market power.

Of the factors mentioned above, the following facilitate a high level of competition:
# Overcapacity
# Low industry growth
# Supplier dependency
# High fixed costs

The following factors limit the level of competition:
# No close substitutes
# Moderate Concentration
# Insignificant exit barriers
# Dispersed buyers
# Entry barriers due to R&D costs
# Product differences.

In spite of these factors the car industry is highly competitive. The reason for this is that the factors that limit competition only provide marginal market power. For example the concentration and product differences that are large enough to provide manufacturers with a little discretion over prices, and a slightly declining demand curve, but too small to provide any real monopoly power.

The factors that facilitate competition on the other hand are very powerful. Especially the overcapacity has led to a very competitive environment, as all manufacturers try to increase their sales through lower prices in order to use their capacity. Thus all in all we will characterise the European Automotive industry as highly competitive.

Nadarasa Subramaniam 1999, Copyright ©, All rights reserved.