1. INTRODUCTION
In recent years there has been a good deal of attention given to domestic aviation policy in Australia. Most of this has concentrated on the advantages and disadvantages of the particular form of regulation which has been in place over the past few decades, the Two Airline Policy. Economists have generally been critical of it, and have pointed to the ways in which this type of regulation blunts the incentives for efficiency. Lately, criticism has become more widespread, and deregulation is no longer regarded as an impractical curiosity out of economists' textbooks. Deregulation of other markets in Australia, along with US airline deregulation, has helped make this possible. Complete or partial deregulation of domestic aviation is now regarded as a serious option.
The focus of this paper is not on the arguments for and against deregulation. Rather, it is on how deregulation might proceed, what its consequences are likely to be, and what problems are likely to emerge if it takes place. To this end, a number of specific questions are considered.
- The first concerns the relevance of US airline deregulation. This has been closely studied, but to what extent can the results be carried over to Australia? It is suggested that there will be important differences.
- The second concerns the working of airline competition. Recent work on US airlines and on analogous markets in Australia enables an assessment of how well competition will work in specific airline markets. It will work better in some (dense) markets than others.
- Third, labour market issues are examined. Deregulation of airline product markets will have a major impact on airline labour markets; this impact can be altered by the nature of labour market regulation.
- Fourth, the alternatives to complete deregulation, some of which can be regarded as stages on the road to it, are examined briefly. In particular, an option which has been given little attention in Australia, the development of a competitive charter sector of the industry, is considered.
- Finally, the problems of moving to a new environment, in which there are gains overall but in which some groups are likely to lose significantly, are examined. Whether compensation should be, and can be, paid to those who lose is a key practical question for a government considering deregulation. The Two Airline Policy has been in place for a long time, and attempts at significant liberalisation have to date been successfully resisted. Some groups perceived that they would lose if it were changed, and used their power to block change. If any change is to take place, either such groups must be overridden or their support for change somehow achieved.
Any work on this subject invites comparisons with the May Committee's Report. (41) In fact, the focus of this paper is somewhat different from that of the May Committee, which has relatively little to say on the five central problems considered here. For example, the Committee is ambivalent about how it expects competition to work out. It considers charter services in passing, but does not examine the possibility of charter operations forming a major sector of the market. Its focus is on describing and analysing the present system and comparing it with some alternatives, rather than on the problems of moving to a new environment. Thus this paper is neither a review of, nor a direct competitor with, the May Committee Report, though in places the same issues are touched upon.
2. WHAT IS WRONG WITH THE TWO AIRLINE POLICY?
The Two Airline Policy has not served Australia well. This is the conclusion emerging from a wide range of independent studies over the past ten years. The policy is a form of regulation which, on a priori grounds, could be expected to create a number of problems. Empirical analysis confirms that these problems are present. While some problems may never be entirely eliminated, they can be much reduced if alternative structures are adopted for the airline industry. One may debate which of the problems are most severe, but several distinct problems can be identified. Five are considered here.
(i) COSTS AND AVERAGE FARE LEVELS TOO HIGH
There have been a number of studies of cost levels of Australian airlines, as compared with those of overseas airlines, all of which indicate that the Australian levels are higher than they need be. An adequate comparison would look at the cost of producing outputs, and relate this to input costs (e.g. labour costs, fuel costs). Two studies which have done this are those of Mackay in 1979 and Kirby in 1984. These indicate that Australian costs were 25-55 per cent higher than US costs for comparable services. (42) The main reason for this is that Australian airlines use considerably more labour than comparably-sized US airlines. (43) Airlines often assert that "special factors" are the reason for these differences -- but these "special (actors" can work in both directions. Good flying weather and lower airport congestion in Australia compared with the US should result in lower costs. It is unlikely that differences of this magnitude can be explained by less efficient labour. Australian airlines do perform better than some (e.g. the smaller European ones), but they are not up with the best.
As a result, average air fares are higher than they might be. With the current (March 1987) exchange rate, economy fares in Australia look low compared with US coach (i.e. economy) fares. This is often quoted by the airlines, but it is an invalid comparison. To tell whether fares are low compared with what they could be, one needs to look also at input costs. At current exchange rates, these are now much lower than in the US. A study which allows for this was published by the present writer in 1985. (44) It suggests that, after appropriate corrections, economy fares are lower than US coach fares, though not by as much as it seems. However, coach fares are used only by a small proportion of travellers in the US, and the average fares paid would be lower than in Australia. All thorough studies of cost, productivity and average fares suggest that they could be lower in Australia if efficiency levels of the best overseas airlines could be reached.
(ii) IMPORTANT MARKETS NOT SUPPLIED
The exact quantitative importance of the previous problem can be debated. However it is clear that the Australian airlines have little interest in supplying the low-fare leisure markets. They supply the business market well, but do not offer the product suited to the leisure traveller. In the US and Europe, fares which are much lower than scheduled economy fares are readily available (45) and leisure travellers make use of these. The same is true of flights to and from Australia. The result is that there is considerable potential for more people to travel by air. (46) This will be fulfilled as appropriate fare types are offered. These lower fares do not come about through magic -- they are the result of higher load factors, greater seating densities and, sometimes, greater efficiency spurred on by competition. Some low fares do exist, but they are subject to tight restrictions and their availability is limited. The airlines find it easier to concentrate on the business market.
(iii) AIRLINES SLOW TO INNOVATE
This is related to problem (ii). It is especially important with regard to innovation in fare types. The airlines have brought in new types occasionally, but this is often at the instigation of the government, not as a result of pressures in the market place. Experiments with different qualities of service have been few and far between.
(iv) INCONVENIENT AND WASTEFUL PARALLEL SCHEDULING
One might think that, in a tightly regulated system, it would be possible to eliminate the wasteful practice of parallel scheduling. In fact parallel scheduling is a product of that regulation -- since most variables, such as overall capacity, fares and routes, are regulated to be the same between the two airlines, there is little scope for schedules to differ. Now that the airlines are flying different aircraft types there is some lowering of the incidence of parallel schedules. This problem may never be eliminated entirely whatever the system of regulation or competition; but it is noticeable that Australia is the only country where the problem is so severe.
(v) OBVIOUS WASTE AND INEFFICIENCY
A tight and restrictive system of regulation often leads to examples of obvious waste. Several come to mind -- one arises with denial of access of QANTAS to domestic routes, to protect the domestic airlines. As a result, QANTAS is forced to fly aircraft between its Australian gateways with poor passenger loads, even though it could easily sell the seats (e.g. to overseas passengers). This in turn limits its ability to compete with its international competitors.
Many economists would suggest that these cases are typical of a regulated monopoly or duopoly, which the Two Airline Policy creates. Insulation from competition means that the pressure to innovate and keep prices down is missing, and the limits on profitability mean there is little scope to profit from efficiency or innovation. It is difficult to "patch up" the system to improve efficiency whilst keeping it basically intact. For example, the introduction of the Independent Air Fares Committee appears to have had virtually no effect. It is not easy to find important general benefits which the Policy conveys, beyond a very high degree of stability. It has created benefits for some groups however -- airlines and airline employees gain at the expense of airline users.
In many cases, for example the US, New Zealand and the Netherlands with their airline industries, and Australia with its financial sector, governments have become dissatisfied with performances under regulation. As a result, they have tried partial or complete deregulation. In all of these cases, the industries have survived, and usually the overall performance has improved. One of the most quoted, and most relevant, examples is that of the deregulation of US airlines, and it is to this that we turn.
3. LESSONS FROM AMERICA
There are many studies of the US experience with deregulation of airlines, (47) including the other paper in this volume. The US case is important because it gives us an insight into how deregulated airline markets work. It does not necessarily follow, however, that deregulation of airlines in Australia would have the same effects as it did in the US. This is simply because the starting points -- the structure of regulation -- are quite different. Some of the major aspects of US deregulation are examined briefly here. They are not given the same relative emphasis as they would be in an American discussion; rather, those changes that might be important in Australia are given greatest emphasis. The most important lessons are considered last.
SUBSTANTIAL NET BENEFITS, BUT WITH LOSERS AS WELL AS GAINERS
This is pretty much in accord with the predictions of economists. (48) Several estimates have been made of the gains in net benefits. (49) All have suggested that there have been major overall gains, which have come from improved cost efficiency, greater availability of different qualities of service and fares, and more frequent services and greater convenience. In fact, Morrison and Winston find the service convenience aspect to be very important (in contrast to the popular perception of the importance of lower fares). Some groups have lost -- notably some skilled labour, and, to a lesser extent than feared, some residents in small towns.
NEW LOW FARE MARKETS
In the US, as in Australia, regulation gave little incentive to airlines to serve other than the full-fare, business market. Deregulation has meant that a wide variety of discount fares have become available, especially on the busy routes. (50) People have taken advantage of these, and this has been a rapidly growing segment of the market. The large majority of travellers now travel on discount tickets. Some of these fares have been offered by low fare specialists, like People Express, but the most important source has been the major trunk airlines, which have been developing this side of their business.
HUB-AND-SPOKE NETWORKS
Before deregulation, airlines could not choose which routes they would fly. They were restricted, in the main, to the networks which had developed over forty years. In this respect, they were quite similar to Australia's East West Airlines at present. When they could choose their networks, they opted for hub-and-spoke systems, which were operationally simpler, and convenient for passengers. It has been estimated that these changes were quite important. (51) They are unlikely to occur in Australia, for the simple reason that networks have not been regulated under the Two Airline Policy, and the airlines have already chosen the mosl appropriate networks.
MINOR IMPACT ON SMALL TOWN SERVICES
One of the worries prior to deregulation was that it might induce airlines t0 drop service to many small communities altogether. While a few towns did lose service, most did not. (52) Jet airlines did drop services, but commuter airlines replaced them, very often with higher frequencies and similar fares. In Australia, this process has been going on for many years, so deregulation would be unlikely to make much difference, though it would probably speed it up.
MAJOR CHANGES IN AIRLINE LABOUR MARKETS
Prior to deregulation, the skilled categories (pilots, flight attendants) of the labour force were mostly unionised, and those who were not unionised tended to be paid at union rates. Deregulation severely weakened the power of unions in the industry. New, non-union, airlines started up, paying significantly lower rates. These airlines then competed with airlines paying union rates. Some unionised airlines made losses and lost market share. Others repudiated union contracts and re-formed themselves under bankruptcy law, offering lower rates of pay -- this was the case with Continental. Some of the more successful larger airlines negotiated pay reductions or agreed to distinguish between old employees (on high rates) and new employees (on low rates).
The impact on the labour market was important for its own sake -- there was an unexpected redistribution of income from airline employees to travellers. Had the airline unions realised how serious the consequences would be they might not have acquiesced in deregulation. However, there is a further important consequence. It resulted in competing airlines paying different amounts for their labour. This has meant a degree of instability for the industry, for example by giving a temporary advantage to some airlines even though they are no more efficient. Stability in competition in the product market can be achieved only when stability is present in the labour market. This suggests that it is important to answer labour market questions when considering deregulation for Australia.
SAFETY UNAFFECTED BY DEREGULATION
In the US, accidents have continued their falling trend during the deregulation period. (53) It can be hypothesised that a competitive airline industry might either pay more or less attention to safety. It seems that the changes that US deregulation involved have had no effect as yet on airline safety. This need not be true elsewhere -- abolishing the regulation as it applies in Australia might lower, or for that matter, raise, safety. However, the US experience suggests that economic regulation does not have much effect on safety.
NEITHER PERFECTLY COMPETITIVE NOR PERFECTLY CONTESTABLE
Perhaps the most relevant aspect of US airline deregulation for other countries is that it gives some insight into how unregulated airline markets work. In broad measure, the results were as forecast by economists. However, there were some aspects and trends which were unexpected.
It was not considered that on each individual route there would be many competitors. Rather, it was believed that numbers of competitors would rise with market density, but that fares on low density routes, with only one or two firms serving them, would be kept at competitive levels by the threat of new entry. Airlines were held up as an example of a contestable market (54) -- and early empirical work tended to support this. (55)
Since then, more work has been done which casts doubt on these views -- in short, free entry and the threat of entry do not seem as strong a force for keeping fares and costs down as was first thought. (56) On routes with low density, fares have risen, and incumbent airlines sometimes seem to be taking advantage of their monopoly power. This power can be quite strong if, as happens in several cases, the incumbent has preferred access to a key airport. To this extent, the analysis of the nature of competition at the individual route level has been revised.
Another revision comes in the assessment of the importance of networks. Previously it was considered that a small airline could compete on equal terms, on a specific route, with a large airline. The advantages that a larger airline has through a wider network appear, however, to be significant. In attracting customers, the larger airline has some advantages and no disadvantages. Unless it is paying lower wages -- and this advantage is being whittled away -- the smaller airline will not be able to survive. Thus, there has been much merger activity in the past few years, and it is believed that the US industry will be dominated by about half a dozen large airlines soon. In addition to network advantages, there are economies of scale in computer booking systems.
Changes are still taking place in the competitive structure of US airlines. Few small, new-entrant airlines have been successful, in spite of temporary cost advantages. The industry may settle down to being one with a group of large firms surrounded by a fringe of small firms. These larger firms will possess and exercise market power. However, they in turn will be supplanted if smaller firms are more efficient or innovative. The pressures for good performance will be greater than under regulation, though not as great as in a perfectly competitive or contestable industry.
4. AUSTRALIA AND AMERICA: DIFFERENCES AND SIMILARITIES
Deregulation in Australia may not yield the same results as deregulation in the US because there are a number of differences in the airline systems and the environment in which they operate. It would be possible to list very many differences, as well as some basic similarities, but four are worth concentrating upon.
THE STARTING-POINT
The Two Airline Policy is a quite different form of regulation from the pre-deregulation US system. As a result, the impact of deregulation is likely to be different. Briefly, the US system regulated entry to routes, and, to an extent, into the industry (though there were many firms there already). It regulated fares, and through these, there was an attempt at regulating profits. Except rarely, it did not regulate the amount of capacity on a route or overall. Routes differed -- on some, there was only one airline, on others many. It was difficult, until immediately before deregulation, to offer discount fares, and there was little incentive to do so. Schedule competition, whereby airlines scheduled additional capacity as a competitive weapon and drove down load factors, was a characteristic of this system. (57)
In Australia, capacity is regulated, and load factors are high (though they are not exactly comparable with US load factors because Australian airlines more often cancel poorly-booked flights and this raises load factors). Fares are regulated, though there is more scope to offer discount fares (though equally little incentive). Profits are virtually guaranteed, though they are loosely controlled. Entry by new airlines is virtually impossible. In the US, airlines had an incentive to be cost-efficient since fares were regulated on an industry-wide level -- it was possible for an efficient airline to keep its profits.
In Australia, the industry consists of two very similar airlines -- while in theory it would be possible for one to be much more profitable than the other, this would invite changes in policy to the detriment of the more efficient airline. The evidence is that the Australian airlines are less cost-efficient than comparable US airlines.
The upshot of this is that the nature of the gains from deregulation could be quite different. In the US, networks and service changes were important -- in Australia they are likely to be less so. In the US, widening of the range of discount fares was significant -- in Australia, the range may not widen very much, but the emphasis on individual fares could change; i.e. low fares may become more freely available. The lower efficiency of Australian airlines would mean that the fall in real average fares could be greater. There is not likely to be the same gains from increased load factors.
THE ATTITUDE OF VESTED INTERESTS
In the US, the opposition to deregulation was not strong. Most groups with any say in the decision were either in favour of it, or ambivalent. The airlines had mixed feelings -- some opposed it, while others supported it. The airline labour force was not strongly opposed to it, once guarantees about jobs had been given. Consumer groups welcomed it. Small communities were wary of it, but they were placated by guarantees of service. The regulating authority, the Civil Aeronautics Board, was basically in favour of it. In the event, some groups, such as labour, lost out -- had they foreseen the full consequences of it, they might have opposed it strongly.
In Australia, deregulation is likely to meet with strong opposition from those unions which represent the airline labour force and some opposition from the major airlines. Smaller communities, and their representatives, are also likely to be doubtful about it. While other groups (e.g. tourism groups) are strongly in favour of deregulation, the existence of strong opposition which is strategically placed will make deregulation very much more difficult to achieve than in the US.
THE INDUSTRIAL RELATIONS SYSTEM
One of the major impacts of the US deregulation was on the labour force, some sections of which lost heavily from it. These changes also had a substantial influence on the way changes worked themselves through -- for example, two classes of airline developed, with one paying lower wages than the other. In the US, there is company bargaining and a fairly low level of unionisation -- in Australia there is the arbitration system, industry-wide bargaining, and very high unionisation. The same changes may not occur. These labour force issues are very important, and they are given specific consideration later.
THE SMALLER MARKET
The US airline market is obviously much larger than the Australian market. The latter is rather similar to some intrastate markets, such as California or Texas. However, it is not clear whether size of overall market is particularly important. Australia has a group of routes which are quite dense by US standards, and which can support a number of competing airlines. The US has dozens of airlines, some of which are very much larger than Ansett or Australian; under deregulation, Australia would have fewer and smaller airlines, but enough for effective competition.
Airline size as such has little effect on performance -- once a minimum size is reached (say 3 to 5 aircraft) a small airline can be just as operationally efficient as a large one. There are advantages to offering a system-wide service, as American and United do in the US, and this helps explain the recent mergers. Ansett and Australian offer a system-wide network in Australia; a much smaller airline than United can gain these system-wide economies in Australia (Ansett and Australian are larger relative to the total market size than any US airline is).
5. HOW AIRLINES COMPETE
It was suggested above that the US experience had raised questions about how competitive airlines will be in a deregulated market. Other countries' experiences also give rise to queries. In New Zealand, airlines have been deregulated for more than two years, but as yet there are no signs of an explosion of new competition. It is worthwhile examining in greater depth how airlines do compete, in order to gain more insight into the likely consequences of deregulating the Australian airlines.
Before turning to this, it is worth noting how an industry with similar characteristics, which is already deregulated in Australia, is performing. This is the long-distance coach industry. Coaches are basically aircraft that don't fly. Most of the characteristics of the airline industry which are quoted as reasons for regulation -- such as the perishability of the product, the mobility of the capital stock and the importance of frequency -- are equally characteristic of the coach industry. Within Australia inter-state, and some intrastate, coach services are not regulated.
The performance of the industry has recently been examined by the Bureau of Transport Economics. (58) Competition, measured by number of operators, has been growing on intercapital routes, and fares appear to be responsive to the number of operators, (59) especially if there are low-fare. new entrants. The busiest routes can support as many as six operators. Operators with nation-wide networks, including intrastate services, charge the highest fares, and compete effectively with lower-fare operators with networks confined to the capitals. Some regional operators also operate on intercapital routes adjacent to their region. High fares spurred new entry, not all of which was successful. It is possible that the older operators will be able to reduce their costs over time, and compete more effectively with new firms. The industry appears to be an efficient one, with fares near minimum practicable levels.
A similar situation could emerge with airlines. There are differences though. The coach industry probably has a much more homogeneous market than the airline industry, and as a result its price structure can be simpler. However, it is clear evidence that competition can survive in an industry with those characteristics of the airline industry which are often thought to give rise to problems. In the following sections, the working of competition in the airline industry is examined. First, competition on a single route is considered, then network aspects are examined, and finally consideration is given to whether specialist services can survive.
COMPETITION ON A SINGLE ROUTE
Economies of scale (or, perhaps, economies of density) exist at the level of the individual route. Frequency is valued, but its price is smaller aircraft, which cost more per passenger to operate. The more dense the route, in terms of passengers per day, the more flights that can be operated with a given aircraft size, or the larger the aircraft that can be used. In addition, there are minimum costs incurred in operating a route. This scale economy exists for the market as a whole, not necessarily for any individual airline. It would be possible for a pair of cities to have five flights per day operated by five different airlines. This is unlikely -- airlines find it is convenient for passengers to operate a schedule with several flights per period. A city pair with only five flights per day would be unlikely to support more than two airlines, though busier markets could support more.
The degree of competition may have implications for the level of prices charged. In a perfectly "contestable" market, this would not be the case, since the threat of entry would be sufficient to force even a monopoly to adopt competitive pricing. (60) Contestability rests on a number of assumptions, not all of which are met in the airline context. The most important deviation from these assumptions comes with the response of an incumbent firm to entry by another. The assumption the contestabillty literature makes is that it holds its price for a time: this is sufficient to allow the new firm to gain a foothold in the market. It has been suggested that this is the "nicest possible response" that a firm could make towards a new competitor. (61) In the event, it is unrealistic in the airline context. Firms can, and do, react to new entry or fare changes by competitors almost instantaneously. (62) This means that they do not have to keep fares at competitive levels to ward off entry. It also implies that if the incumbent airline has some advantages (e.g. if it is better known) it will be difficult for a new airline to stage a successful entry.
The upshot of this is that on routes with low density, the existing airline will have some monopoly power, and it will use it. The threat of entry is insufficient to control the monopolist. Prices will fall if a firm attempts to enter, but this will be a temporary phenomenon; they will rise again when one airline is successful in pushing the other out of the market. The actual amount of competition, measured by the number of firms, is a determinant of fare levels. (63) This also explains the persistence, since deregulation, of price discrimination. Fares are often set to segregate different groups of consumers, and differences in fare levels do not always accurately reflect differences in costs. Price discrimination is a characteristic of monopoly, not competition -- if it exists in a competitive environment, firms will find it worthwhile undercutting those fares which are above costs. Certain fare types which persist in the US can be regarded as evidence of price discrimination, since they cannot be justified on cost grounds.
Allowing free entry on to a route may not be as strong a discipline as was imagined if entry does not take place, but it is still likely to be worthwhile. First, it will be possible for a new firm to supplant the old one if the latter is inefficient. Fares may be cut when the new firm appears, but not low enough to stop the new firm succeeding. Free entry will not guarantee that monopoly power is not exercised, but it will ensure that the most efficient firm is the one which supplies the market. Secondly, while firms will not behave in a perfectly contestable manner, a firm may still take account of potential entry when setting fares, if only because it wishes to avoid provoking a price war, which will be expensive even if it wins. (64) It is an empirical question how strong potential competition will be as a force disciplining market behaviour. The evidence suggests that the effective monopoly power is not strong.
THE ECONOMICS OF AIRLINE NETWORKS
One of the lessons from the US experience is that airlines with good networks appear to have an advantage. (65) This has been accentuated with the development of hub-and-spoke networks which already operate in Australia. The logic of these networks is simply that it is often cheaper to fly people through a hub than to fly directly between two points using smaller, more expensive aircraft. Many trips in Australia, other than between adjoining cities, are made through intermediate airports: Canberra-Adelaide via Melbourne or Port Hedland-Sydney via Perth, for example.
There are distinct advantages, in terms of coordination and convenience, in a passenger making the two stages of the trip on one airline. An airline which is unable to offer through service will be at a competitive disadvantage compared with one which can. This is a point strongly made by QANTAS in discussion of its disadvantages vis-à-vis US airlines on the Pacific route. These airlines can offer a whole range of US cities, whereas QANTAS is restricted in what it can offer in Australia. Through service can be offered with agreements between airlines, though the scope for this is often limited -- firms do not wish to make commercial agreements with firms with whom they are competing on other routes.
Operating a coordinated network may be more expensive than a pattern of uncoordinated services. Schedules must be arranged so that it is possible for passengers to change flights. It is difficult to know whether this constraint on scheduling is likely to be costly or not; coordinated schedules may be desirable to obtain good utilisation of aircraft. US experience, which is supported by Australian experience, suggests strongly that any costs of coordination are outweighed by benefits on the demand side.
The airline with the single flight or simple network will thus be at a disadvantage vis-à-vis the larger one. Through passengers will prefer the larger airline, and so, other things being equal, the smaller airline will attract less passengers on a given route. It will either have lower load factors, which will entail higher costs per passenger, or lower frequency, which will make it a less attractive alternative. It may be possible for a new airline to enter with a few services if its costs are lower than those of the incumbents -- it can then offer lower fares as an enticement to passengers. Such cost advantages are likely to be temporary, and it will not be possible to stay small and survive.
THE SCOPE FOR SPECIALIST SERVICES
One way in which a small airline might survive in competition with large ones might be to offer specialised services which appeal to a niche in the market which is not well served by the others. The scope for doing this is, however, quite limited. In the US, for example, there are not many specialised smaller airlines. The charter operators, which specialised in low-quality leisure travel, were almost eliminated by dereglation.
The reason for this is that there are economies of scope in airline services. It is cheaper to supply a range of different qualities, in terms of toth in-flight service and booking conditions, in one aircraft than in several. Through APEX fares, which commit the passenger well in advance to travel on a particular flight, the scheduled airline offers a charter-type service and fills up capacity at the same time as it offers an on-demand, convenient service for the business traveller. Only where it is not possible to mix travellers on the same aircraft is it necessary to use separate aircraft. This might be the case if the objective is to supply a low-quality service with very high seating densities. It is worth noting that where specialist services exist, as they do in the UK and Europe with charter airlines, the segregation of airline types is brought about by regulation -- the different airline types are not permitted to compete directly with one another.
This is not to argue that specialist services are impossible -- rather, they are unlikely to be an important way for smaller airlines to compete effectively. A small airline might be able to develop a secure market through product differentiation, especially if the larger airlines ignored some market niches. Thus, it might sell actively in ethnic markets, to foreign tourists, or to older people, and tailor its product appropriately. The first to tap these markets might gain a long-term advantage which would be difficult for other airlines to compete away.
There is one important form of specialisation which can provide a strong market base -- regional specialisation. An airline which is based in a particular region and serves it comprehensively can operate flights to and from that region in effective competition with national trunk airlines. It will have strong local feed, even though it will not have good connections beyond the region. It will obtain a good share of regionally-based passengers on flights to or from the region, though a smaller share of passengers who wish to make onward connections. The regional carrier, USAir, based in the north-east of the US, has been able to use its regional feed effectively so as to compete on long distance flights from its hub.
CONCLUSIONS
The analysis presented in this section might suggest that an airline system is likely to become monopolised. Most of the advantages lie with the large airline with the extensive network (and computer booking systems). The recent merger activity in the US, and the growing dominance of a half dozen or so large airlines, seems to confirm this impression.
However, there is scope for some competition to exist and have an impact if there is free entry. There are other factors which have not been given detailed attention. There can be problems in coordinating a large airline efficiently, and even if they are solved, this may be done in a way which lessens its ability to respond to changes in its environment. Smaller airlines may be more efficient, and, perhaps more significantly, some of them may be more innovative; they will then have a chance to supplant the larger airlines. This does happen in oligopolistic industries, dominated by a small group of large firms and a fringe of small firms. The more flexible nature of the capital stock, as compared with manufacturing capital, should mean that it is easier for firms at the fringe to grow and become large themselves. This is especially true if some retain a regional base.
6. HOW DEREGULATION WOULD WORK IN AUSTRALIA
The experience of the US and other airline markets, along with that of deregulated markets for other services, gives us a good idea of how a deregulated airline market would operate in Australia. Several aspects are considered here. First, we examine where the competition would come from. The criteria for successful entry are next considered. The fortunes of the major trunk airlines in this environment are then examined. Fourth, the implications for different routes, and fifth, pricing behaviour of the airlines are discussed. This enables us to conclude who would be the likely gainers and losers.
THE COMPETITORS
It may be taken for granted that the two major trunk airlines, Ansett and Australian, will be present in the new environment. So too will East West, which will have the scope to expand. East West's position may not be particularly secure, however, as it will be open to competition on many routes on which it currently has a monopoly. In the initial stages, before new airlines emerge, the major airlines may concentrate their competitive thrust against it. However, East West does have the advantage of a strong regional network, and it could use this as the basis for selective expansion, as USAir did. Commuter airlines may expand their operations, but they are usually small, with access to limited finance, and are often linked to the major airlines.
The other immediate competitors are the international airlines -- most likely QANTAS, and perhaps Air New Zealand. These airlines could easily operate on trunk routes -- indeed QANTAS flies several already. They are relatively efficient airlines, and they could add a considerable amount of capacity to domestic routes in a short period. They have built their recent growth on innovative pricing and leisure travel, and could be expected to target leisure markets initially. Three problems associated with allowing these airlines to compete are:
- Currently they do not pay the same taxes (e.g. fuel taxes) as the domestic airlines. It should be simple to ensure that they obtain no net advantage from this.
- Their domestic operations would be a small proportion of total operations, and they may cross-subsidise from regulated international routes, thus destabilising domestic routes. This is probably unlikely given the modest profitability of international routes.
- They may not compete actively for fear of encouraging more competition (eg. from Ansett) in their regional international routes.
The most interesting prospects surround the possible development of new airlines. It does not take long to set up a new airline, especially with leased equipment. It is possible that many hopefuls would enter, and that only a few would succeed. However, it seems more likely that there will be only a few entrants. These would ensure they had a basic capital city/tourist areas network, or possibly a regional base (e.g. Queensland plus routes to Queensland). Most likely some would come in at the low end of the market, aiming for leisure traffic. If they possess marketing or cost advantages, their entry could be successful. Inevitably they will find competition from the major airlines difficult, and few are likely to survive in the long run. However, two or three successful new airlines would be sufficient to produce major changes in the industry.
CRITERIA FOR SUCCESSFUL ENTRY
Entry into a deregulated airline market, dominated by two large, experienced airlines, one of which is owned by the government, the other owned by large multinational corporations, is unlikely to be easy. A good deal will depend on how the existing airlines respond. It cannot be guaranteed that their response will always be immediate, and, more importantly, appropriate. Ansett Pioneer's and Greyhound's slow response to new entrants allowed them to gain a foothold in the coach market. (66) Airlines certainly can change fares quickly, though it takes time to alter the nature of a service. The major airlines may make mistakes about markets -- e.g. they may believe that a market is unimportant, and neglect to seek it actively when others are doing so.
In many markets, innovation is the key to successful entry. In the airline market, it can be important, but the problem is that innovation can usually be easily matched. Innovation includes designing a new fare type to appeal to a particular market. There are possibilities for the first-comer to obtain an advantage, especially where airline travel is linked up with other sevices. Thus, if an airline links up with resorts in an area, or travel agents serving a particular market, latecomers may not be able to match it.
A new airline is likely to have cost advantages; this would probably be its most useful competitive weapon. One obvious cost advantage would occur if it obtained labour at lower wages than existing airlines, though this would be temporary, since the older airlines would be able to reduce wages to the market level over time. However, the cost advantages are likely to exist for a time even if all airlines pay the same award wages. The evidence is that the Australian airlines are less efficient than they might be, especially in their employment of labour. While still paying award rates and adhering to award conditions, a new airline would be able to operate with less labour-intensive work practices and introduce labour-saving technology from the start. It would take the existing airlines some years to negotiate improvements to match them.
The successful new entrant will almost certainly have an orientation towards the leisure market. This is partly because that is the market which is currently not well served, and also because the major airlines have several devices, such as frequent flyer programmes, which they could use to retain business traffic. To tap the leisure market effectively, it would be necessary to form links with travel agents and resort operators -- in the way the charter airlines in Europe do. Entry is thus made more difficult by the fact that many of these already have links with, or are owned by, the major airlines.
Finally, while it is comparatively cheap to set up an airline, it can be expensive to operate it. Strong financial backing is necessary and successful entrants are likely to be subsidiaries of larger companies. This is for two reasons. First, if there are price wars as the major airlines respond to the entrants' initiatives, it may be necessary to incur losses for a considerable period. Second, if entrants are known to have financial backing, the major airlines will be less inclined to indulge in price wars which are expensive to themselves, and will more willingly allow the entrants a part of the market.
THE MAJOR AIRLINES
As indicated above, the two major airlines have many advantages compared with their competitors, and they can be expected to survive. At least one will prosper, but possibly not both. If one is more innovative and responsive, it will grow at the expense of the other; the current balance of equally-sized airlines is artificial. This has important implications for the government's control of Australian Airlines. If the latter is to succeed in the new environment, it must be given full commercial and purchasing freedom. If decisions must be approved by controlling Ministerial departments, they will be delayed, and the airline will slip behind its competitors.
Currently, the major airlines can do what they want so far as services, schedules and fare types are involved. They are subject to price control, and, through this, a weaker control of profits. Over time, Ansett has earned higher profits than Australian, and has earned profits which are good, but not remarkable, for the type of industry. Recently its profit performance has been very good, partly because price control has not been tight. Deregulation would mean that profits would be more volatile, and for Ansett, they would possibly be lower than at present. The airlines would be subject to more competition in profitable markets (e.g. Sydney-Melbourne) but would also be able to take more advantage of monopoly situations, and be able to profit more from innovation. It is always difficult to compare profitability under a regulated and a deregulated environment, since regulation can be used to raise and stabilise profits above competitive levels; this could happen if the Two Airline Policy continues. In a deregulated environment, successful airlines would earn, over the long term, at least what can be earned elsewhere in industries with comparable risks.
COMPETITION AT THE ROUTE LEVEL
It is worth making a distinction between three types of routes. First there are the dense routes, such as the intercapital and major resort routes. The largest fifteen of these accounted for 65 per cent of total airline passengers in 1983-4. (67) Next there are the less-dense routes operated by the major and regional airlines. Finally, there are routes served by commuter airlines. There are no rigid distinctions between these categories.
There are unlikely to be major changes on commuter routes. In some cases, where the routes are quite dense (e.g. Sydney-Newcastle), there is the possibility of additional competition, especially if it fits the network of a larger airline to have some feeder routes. Most change is likely to take place on the busier routes. On the long-haul routes (e.g. Melbourne-Perth), and those linked to tourist gateways, QANTAS (and any other international airlines permitted) may operate, initially on the basis of carrying international tourists and filling up capacity.
On busier, shorter-haul routes, most of the competition is likely to come from new or expanded regional airlines. On one or two routes there may be as many as half a dozen airlines (e.g. Melbourne-Sydney), though a few will have the major share of the business. On others, there are likely to be three or four competitors: US experience suggests that this is a large enough number to ensure that fares are kept low and collusion is difficult. On the busiest fifteen or so routes, which account for two thirds of total traffic in Australia, fares can be expected to fall, and the range of fares offered will increase. It is here that the main benefits from deregulation will be concentrated.
On the less-busy routes, competition is likely to be limited. One or two airlines will usually operate, and on many routes potential competition will not be much of a threat. Some of these routes are currently cross-subsidised, though not by very much; they would cease to be. Where potential competition is little threat, airlines will take advantage of their monopoly position. It is possible that this will be less of a problem than in the US, because of the geography of Australia and the shape of its networks. Most low-density routes operate out of a large hub (a capital city) into which several airlines fly. It will be easy for other airlines to challenge incumbents on monopoly routes -- in the US, route structures often make it awkward to enter low-density routes out of minor hubs.
PRICE COMPETITION
Price competition is likely to be strongest on the busy routes. It will take the form of lower prices for the same service, and the introduction of new types of services and fares. The major airlines, with extensive services and a reputation for quality, will be able to charge slightly higher fares than new airlines. It is possible that full fares (economy and first class) will not change by much, even if average costs are reduced. This is because they will be appropriate for the business traveller, and airlines will not find it worthwhile to discount them. Average fares will fall, with a greater proportion of passengers travelling on discount fares.
Price wars will inevitably happen, though they will not be chronic (unless the government subsidises losses on the domestic operations of Australian and QANTAS). Price wars emerge when a new operator is willing to sustain temporary losses to establish itself on a route, or when there is excess capacity. Thus, in the US, periodic price wars on the transcontinental route occurred when airlines had surplus long-haul aircraft. (68) Probably long-haul routes are more susceptible than short-haul ones because there will be more airlines with appropriate aircraft, (because international airlines like QANTAS may be active on them), and periods of excess capacity in international markets will not coincide with such periods in domestic markets.
Predatory pricing is likely to be a problem at times. The May Committee gave considerable weight to this possibility. (69) The larger airlines will have the ability to incur losses on some routes to force new entrants out. There is some debate as to whether predatory pricing is rational for a profit-oriented firm; however, several examples in which predatory pricing is rational, for example to establish a reputation as a tough competitor, have been given. (70)
It may also occur because incumbent firms make mistakes, and fail to act in their long-term interest. Predatory pricing is very difficult to prove, and often legal remedies are too late. It is also virtually impossible to stop through regulation (e.g. on minimum fares). One possible option is to allow fare changes, in either direction, only after a set period of operation. The problem is that this would be difficult to enforce when the airline is able to introduce new fare types, which can enable it to lower fares when competition arises. Predatory competition is a possibility for all competitive industries, especially those with a few large firms, and it is an empirical question how serious it would be in domestic airlines. If it is a serious problem, possibly the most effective control is through severe penalties, which would act as a deterrent even if the probability of successful prosecution is low.
THE GAINERS AND LOSERS
The main gainers from deregulation would be leisure travellers on inter-capital and tourist routes. The gains would be in the form of lower fares and expanded opportunities for travel. Business travellers would not gain as much (unlike the case in the US), but they would gain from reduced standard fares on some routes (especially long-haul), greater frequency, more choice of fare types and a lessening of parallel scheduling. Associated industries such as the tourism industry would gain, as there would be more domestic air travel by home and overseas tourists.
While, on balance, there will be substantial net gains, there will be losses incurred by some. Not every passenger will gain. The small minority who travel on low-density routes may lose through higher fares. The major airlines will survive and at least one will prosper, but they will possibly do less well than Ansett is currently doing under the present loose regulation of fares (which need not continue even if the Two Airline Policy does). Competitors for the airline industry, especially the interstate coach industry, will lose some business -- this may take the form of reduced growth rather than contraction.
The group which is likely to lose most, at least on a per capita basis, is airline employees. This could happen even if the industry remains quite unionised and award wages are enforced. Along with the major airlines, this group sees itself as the main potential loser from deregulation. The labour force issues are quite critical if total or partial deregulation is imposed, and they deserve specific attention.
7. OTHER OPTIONS
Complete deregulation is not the only change that can be made -- changes to the Two Airline Policy can be made short of this. They are many and varied, but they may be grouped into three general headings, which are considered here. One is to move to a three airline policy, another can be described as partial deregulation, and a third is the competitive charter option. Variants of the first and second of these are discussed as options 2, 3 and 4 by the May Committee. (71)
A THREE AIRLINE POLICY
A three (or n) airline policy is often dismissed as being little different in consequences from the Two Airline Policy. (72) Under certain circumstances this would be the case, but not necessarily under all circumstances. Results depend very much on the regulation which applies under the three airline policy.
The key control is capacity control. If there are to be three equal airlines, or two major airlines and a third of predetermined size, capacity control will be needed to preserve the chosen balance. It may be that a third airline is permitted to grow in a controlled way so that eventually it obtains major airline status. Under the Two Airline Policy, neither airline has an incentive to compete strongly for more traffic because it will not have the capacity to handle it. The same would be true under this variant of the Three Airline Policy. It is very likely that, if a restricted third airline were permitted on trunk routes, either on equal terms with the majors, or to provide "specialised" services, there would initially be a period of active price competition. This would only last until the new airline had established its permitted market share -- after this, pricing would revert to normal.
Alternatively, capacity control could be dropped as three airlines were permitted to compete openly on trunk routes. If this happened, long-term changes could come about. Two airlines, not capacity-controlled, could act as a joint monopoly by colluding tacitly. It is more difficult for three, especially if the third has a lower cost structure (at least, in the short term). From the point of view of the major airlines and their employees, this form of regulation might be little better than complete deregulation.
Most likely, if a third airline were admitted to the group, it would be under the old rules. The game could be played again by a minor airline, if it saw East West successfully break into the group. Indeed, East West may be currently accepting poor returns from operating in too many markets with inappropriate aircraft, as a way of forcing itself into the profitable group (i.e. it may be currently indulging in "rentseeking"). There would be an incentive for other airlines to follow its example if it is successful.
If a third airline is to be admitted, which should it be? There are several possibilities, apart from the obvious one, East West. QANTAS has a claim -- as might Air New Zealand, granted that it is now facing competition from Ansett on its domestic routes. It is currently the best equipped to service trunk routes in Australia. Perhaps one of the commuter airlines, such as Lloydair, or a freight airline, such as IPEC, both of which have experience operating jets, have a claim. If innovation and pressure for efficiency are desired, perhaps an altogether new airline would be best. In issuing a third airline licence, the government is giving away a very valuable asset -- ideally it would give it to the airline which seems likely to make the most improvement.
If a third airline does obtain permission for trunk routes, it may be more efficient and innovative than the present two. If so, it would earn higher profits once it was established. This would put pressure on the policy. It would be taking business from the present two (assuming capacity is still controlled), and these, along with their workforces, would be worse off. There would be pressure for the third airline to lower its efficiency (e.g. by adopting the inefficient work practices of the other two). Even if there were no overt pressure for this, it is unlikely that it could remain more efficient and more profitable for long, and its incentive to innovate would be weakened. Entry of a third airline on a capacity-controlled basis could have an impact on airline efficiency, but only a temporary one. Furthermore, it would not have much impact on the two major airlines. Thus, from a national point of view it is not an attractive option.
PARTIAL DEREGULATION
Partial deregulation can mean many things, but one distinction that can be made from the start is between deregulation of part of the system, and removal of some controls over the whole of the system. Both these are feasible options.
One form of partial deregulation would be to remove controls over operations in a specific region (e.g. internal NSW) or over specific types of routes (e.g. interstate routes between non-capital cities). It would be feasible to remove all controls, or only some of them. One possibility would be to allow specific types of services on particular routes.
If improved efficiency is the objective, the largest gains would be achieved by allowing more competition on the busiest routes. Most people travel on these routes, and competition is likely to work best on them. Indeed, if there were a fear of use of natural monopoly on low-density routes, there could be a case for retaining some regulation of them. The impact of such changes depends on how tightly capacity is controlled -- if it is tightly controlled, new entry or flexibility in other areas, e.g. fare setting, will have little impact. If airlines are able to schedule extra capacity, and new firms can come in, the impact could be large. Allowing only specialised types of service will limit the impact and result in an inefficient allocation of business -- this is because economies of scope dictate that it is desirable to have most types of service operated by the same aircraft and airline.
An alternative would be to deregulate the minor routes -- this would be a solution which disturbed the status quo least. The gains are likely to be small, since the traffic is not large, and the possibility of competition may not become the actuality of competition very often. In fact, under certain circumstances, this option could lower the overall efficiency of the system. This would happen if new airlines were permitted to serve direct flights between cities now only served by indirect flights. The new airlines will serve whatever routes they are allowed to, and operate these direct flights where they can. This will lower frequency and/or raise costs on the indirect routes and make them less viable. While this cannot always be presumed, the major airlines had a choice of direct or indirect flights, and they chose the latter, presumably because of efficiency. Allowing restricted free entry could result in a less efficient network ol services.
The second form of partial deregulation involves removing some aspects of regulation across the system -- for example, removal of fares control, capacity control, or controls over entry on to routes. Removal of one form of control need not lead to an improved overall performance. For example, removal of fares control could lead to a situation where the monopoly power created by capacity control was used more extensively. Another possibility is that removal of certain controls might not have much impact -- for example, removal of controls or entry to particular routes might not have any effect if no other airline is permitted the capacity to serve them.
As always, the key control is that over capacity. If capacity is tight, airlines have no incentive to lower fares. Profits may be earned, though over time they may be whittled away by "cost padding". (73) Fares may not be controlled, and entry may be permitted on to routes and into the industry; but, as long as capacity is controlled, fares will be high and costs will tend to be excessive. Capacity control invariably appears to be balancing demand and capacity effectively when it is looked at after the event, because, if it is scarce, fares rise to ration excess demand, and possibly costs will rise and absorb much of the profits.
The impact of capacity controls depends on how the capacity is allocated, and how broad the capacity constraint is. Allocation of capacity is akin to allocation of rights to print money. As long as capacity is restricted, rights to it will be valuable. If it is exactly "right", from the point of equating demand and capacity at competitive prices, it is unnecessary (further, it is highly implausible that any regulator would ever be able to balance demand and capacity correctly in such a rapidly changing industry as aviation). It is easy, though inefficient, to grant capacity in an arbitrary way, on the basis of previous shares, or equal shares to each airline, as under the Two Airline Policy. If capacity is restricted, it is desirable that it should go to the airline which will make the most productive use of it. A system of bidding for scarce capacity will give the airlines an incentive to use it efficiently, and allow scope for entry by new efficient airlines. This would work provided profits are not regulated, either explicitly or indirectly (as at present).
Most of the forms of partial deregulation that are likely to be tried will not make for much change in the nature of the regulatory system. To do so, changes must alter the ease with which capacity is obtained and deployed on routes, and alter the incentives forcing the airline to make efficient use of its resources. The impact of a partial deregulation depends very much on how tightly or liberally the regulators administer the remaining controls. It is worth remembering that the changes introduced to the Two Airline Policy in 1981 were described as major liberalisations in the area of fare-setting, route entry and coordination between airlines. However, they have had little perceptible effect on the operation of the Two Airline Policy, or on the efficiency of the airline system, though fare-setting has become more flexible.
THE COMPETITIVE CHARTER OPTION
The possibility of charter airlines operating within Australia has been given little attention, even though the development of a competitive charter sector would prove a way of achieving some distinct change without putting much pressure on the regulated sector. Overseas, charter airlines are used extensively as a means of ensuring that leisure markets are served without deregulating the scheduled airlines. They operate at lower cost, and have fares very much below those of regulated scheduled airlines (at about one third to one half of scheduled fares). This very large difference might not be repeated in Australia, but a competitive charter sector would ensure that fares were as low as practicable.
There is nothing inherently advantageous in the charter system of wholesaling bulk capacity to travel agents and selling this on a restricted basis (e.g. in terms of changing flights or obtaining refunds). Rather, the charter sector, as exemplified by that in Britain, is a way of segregating markets. Some, admittedly artificial, restrictions are imposed on the freedom of charter airlines to sell tickets. As a result, they cannot compete too directly with scheduled airlines -- they cannot, for example, offer a product which would be attractive to the business traveller. The more restrictive, but cheap, travel is ideal for the leisure travellers who are paying their own fares. The primary distinction between charter and scheduled sectors is not so much the type of service, but rather that, because markets are segregated, it is possible to have quite tight regulation of one sector along with competition in another. Entry on to scheduled services can be restricted, but entry on a charter basis can be free. (74)
While the distinctions between the two sectors are never complete, the two may operate with little interference from one another. So far as the scheduled airlines are concerned, there is little addition to competition. If they are less than efficient, they will remain so. A charter policy would produce less benefits from improved efficiency than would complete deregulation. It would also impose little in the way of losses on any particular groups. The scheduled airlines would not suffer, and their employees would if anything be better off, with the opening-up of job opportunities in a related sector.
In the leisure market, it would be possible to have a moderately competitive and efficient industry. For the leisure traveller fares would fall and the market would be served to the fullest possible extent. If one of the objections to the Two Airline Policy is that the leisure market is poorly served, a charter sector would resolve this. A small number of charter operators (say as low as three) coupled with free entry into the industry would be sufficient to produce quite competitive results. Charter airlines are easy to establish, and can start quite small.
Airlines such as QANTAS, and possibly other international airlines, would be able to make effective use of spare capacity in a charter market. In some cases, e.g. with groups of overseas tourists, the market for charter type operations is already there. The market for domestic internal travel is yet to be developed. Charter airlines are likely to concentrate on inter-capital and tourist routes which are already dense, and they are unlikely to have much impact on the low density routes.
A competitive charter airline system may thus be regarded as a possible half-way house between the Two Airline Policy and deregulation. It would bring benefits, though these would not be as great as those obtainable from more thorough change. Its advantage is that it would be almost costless -- unlike the case with full deregulation, the major airlines and their workforces would lose little if anything. For this reason it should be relatively easy to implement.
This system might or might not be regarded as a step towards deregulation. In the UK the charter system has been developing and significant for more than twenty-five years, and yet regulation of scheduled airlines has changed rather little. With charter airlines it would be possible to observe the workings of airline competition in Australia, and it would be possible to assess its performance. The segregation of the two markets means that firms in one market are protected from competition from firms in the other. It becomes possible for new firms to become established more easily than in a totally deregulated environment. If there are doubts concerning the viability of new entrants in a market hitherto dominated by the two major airlines, a period of charter operations prior to full deregulation would be a means of helping establish new firms.
The charter/scheduled distinction is artificial, and thus it is consistent with inefficiencies remaining. The primary limitation of the segregated market system, that it has little impact on airline efficiency in the scheduled sector, is also its main practical advantage -- it is possible to achieve some gains without adversely affecting groups which will seek to block more far-reaching change. If the charter industry fails to develop, the benefits will have been small because the existing industry will have been serving the market well. If it does develop and become large, the benefits will be correspondingly large, even though less than those from complete deregulation. Evidence on demand and costs in Australia suggests that the gains will be large rather than small.
8. THE LABOUR MARKET UNDER DEREGULATION
The system of regulation under the Two Airline Policy can be argued to have had an impact on the market for airline labour. If this is true, reforming regulation or deregulating the product market will have consequences for the labour market. The way in which these changes are handled can, in turn, affect the way in which deregulation works out. Questions which are major in their own right, such as that of whether labour market regulation is appropriate, become relevant. However, they are only discussed tangentially here.
THE TWO AIRLINE POLICY AND THE LABOUR MARKET
It can be argued that some groups of airline employees have done well out of the Two Airline Policy. (75) The policy creates a joint monopoly which is limited in the extent to which it seeks profits. There is scope for employees who are essential to the industry to push for higher wages or better conditions and for these to be paid for by price increases. Looked at another way, there are monopoly profits in which employees seek a share. The incentive for employers to resist wage demands is weakened by the nature of the regulation. The result is higher-than-market wages and conditions for some airline employees. For example, Australian crewing requirements for some aircraft are higher than almost anywhere else. In a number of areas, wages are above market levels; this is because there is effectively a queue of suitably qualified persons to obtain a job. Econometric studies, such as that of Findlay, (76) indicate that wages for some employees are well above what could be expected from experience and job skills. In the US, the changes in the product market brought about major changes in the market for some employees, such as pilots. While more pilots were employed, they faced lower wages. The evidence is that some groups of employees have shared in the monopoly profits created by the Two Airline Policy.
These gains have mainly been made by those groups which are specific to the airline industry. The airline industry requires pilots; without these it could not operate. It also needs computer operators -- but it employs only a small fraction of the total employed in the economy. If a group of employees can effectively restrict or control entry into the industry, and they are essential for operation of the industry, then they will have a strong negotiating position. Pilots will be in such a position, whereas office staff will not.
Individuals will undertake considerable training and sacrifice in order to obtain a position as a favoured employee. Pilots will operate, for low pay, for many years, in the general aviation sector in the hope of getting a well-paid job in the major (or regional) airlines. There can be waste -- they have an incentive to pick up more qualifications than they really need. Thus, an individual need not do especially well over his or her lifetime when all is considered. This would be a good example of "rent seeking".
Some groups, such as pilots, probably do well out of many forms of regulation -- they are well paid in many parts of the world. They may preserve their strong positions in a deregulated environment. Other less well-organised groups of employees may not be able to preserve their position if the regulatory system is changed. For many airline employees, and perhaps a majority, regulation may have little effect on their pay and conditions.
THE CONSEQUENCES OF DEREGULATION
Deregulation is likely to bring more (and different) airlines, more traffic, and greater attention to profit and to cost control. This will happen in an industry which is currently highly unionised and in which wages and some conditions are set by tribunals (which may be just enforcing agreements achieved between unions and airlines). Wages are regulated, but their level reflects, at least to some extent, the underlying demand and supply conditions. What happens under deregulation depends in part on how this labour market regulation works. (77)
First we should look to see how strong the market position of a specific group of employees is. This will be correlated with, but not the same as, the degree of unionisation. Deregulation may mean that some sections of the workforce prefer to work outside the existing unions; this will mean that the strength of the latter is reduced. However, it is also possible that new employees may work through the union, but have different objectives (e.g. they could be more interested in job opportunities than increases in pay, or vice versa). It will become more difficult to ensure widespread unionisation, and to achieve it the union will have to make concessions to new members. Some groups (perhaps pilots) may be able to preserve their market position while others may not be able to do so.
Second, we have to examine how labour regulation works. It is possible that the underlying strength of particular groups may be weakened, but their wages may be kept up by regulation. If external standards (e.g. wages in comparable industries, or for the same job in other countries) are used in setting wages, they will not fully reflect changes in the market environment. These external standards are quite important in Australian wage-setting, but they are not used to the complete exclusion of market considerations. It would be possible for regulation to preserve wages and conditions, but such regulation itself would come under pressure from airline deregulation.
In summary, airline deregulation is likely to weaken the market power of some specific groups of employees -- but with some groups being relatively unaffected, and others much affected. The market power of some airline-related groups (e.g. air traffic controllers) could be enhanced. The outcome depends on how rigid the labour market regulation is. It is likely that labour market regulation will be used to an extent to preserve the position of groups which have lost. However, it is improbable that it will be effective in overriding the underlying market changes completely. The result is that, to a degree that depends on different groups, deregulation will lead to lower wages and/or worse working conditions for some groups which currently do well from the Two Airline Policy.
Overall, the airline labour force is likely to expand. As traffic increases, and new airlines appear, there will be a demand for more staff. The established airlines will seek to increase labour productivity, and will shed labour in some areas, though they will take on more in others. Some groups of employees will face improved market conditions, and their wages will tend to increase. The regional composition of the labour force could change (e.g. more may be employed in Queensland and fewer in Victoria). For specific employees on the outside of restricted occupations -- e.g. general aviation pilots unable to get into the airline industry -- there will be more opportunities and higher pay. In the US, even with less unionisation than Australia, overall wage rates did not alter much as a result of deregulation. (78) The same is likely to be the case in Australia, though small groups of specific employees may suffer.
If a group suffers a wage cut, this is neither necessarily a good nor bad thing overall. If it happens, consumers will obtain lower fares. This is a transfer from the employees to consumers. If deregulation produced lower fares solely at the expense of lower wages, its merits would be doubtful. If these transfers do take place, and it is likely that they will to an extent, then whether they are desirable or not depends on society's views as to equity, and the position of the gainers and losers. It may well wish to compensate losers, either on the grounds of equity, or merely to secure their acceptance of changes.
THE CONSEQUENCES FOR DEREGULATION
What happens in the labour market partly determines what happens with deregulation. If there are no changes in wages or conditions (for example, because labour market regulation is effective), the consequences of deregulation will be less far-reaching, but the transition will be smoother. If all airlines must pay the same regulated rates, they will have more similar cost structures (they will differ only to the extent that they differ in the efficiency with which they use inputs). New airlines will be operating with the same opportunities as established airlines so far as costs are concerned, but they will be at a disadvantage so far as networks and familiarity with consumers are concerned.
In contrast, if the market power of specific employee groups is weakened, and labour market regulation is not used to preserve wages, then new airlines may have an advantage. They may be able to hire people at lower wage rates or under more advantageous working conditions than the established airlines. It is not necessary to have "non-union" airlines for this to happen, and it can be quite consistent with industry-wide award rates of pay. For example, promotion policies will be different, meaning that the same person would be paid differently in different airlines. While regulation is used to reduce differences, it is unlikely that it will be completely effective. The newer airlines, or those with more flexible labour arrangements, will have a cost advantage over the others. This will make it easier for them to enter and compete. The established airlines will later be able to match the cost levels of the new entrants, and this will enable them to recapture lost ground. In the long term, the position may be the same for new and old airlines. However, in the all-important period of new entry, the more tightly regulated the labour market is, the more difficult it will be for new firms to enter. The airline labour market in Australia is more tightly regulated than that in the US, and thus the temporary advantage to new airlines will be less. There is a policy decision to be made as to how tightly airline labour markets are to be regulated, and this is one aspect which is relevant to this decision.
9. OTHER ISSUES
If Australia decides to deregulate fully or partially, there will be several subsidiary issues that need to be addressed. Four are considered here -- constraints on the operation of competition, foreign participation in domestic aviation, the ownership of Australian Airlines, and residual regulation.
CONSTRAINTS ON COMPETITION
Competition in domestic aviation may be permitted, but it may not come about for some time. Several constraints may prevent new airlines from operating or expanding as they would wish. Some constraints are likely to prove unimportant. Finance is one of these: setting up an airline does not require much finance, especially if equipment is leased. The Australian capital market is quite able to finance any new airline which is considered a good risk. Expertise -- managerial and operational -- could be a problem for a period, though the airline sector in Australia is quite large, and there are many qualified personnel (e.g. pilots) available to work for new firms. Both Ansett and Australian operate computer booking systems. New airlines (and East West, if it opts out of Australian's system) will have to make alternative arrangements. Other booking systems (transport and non-transport) do exist, and software is available from overseas, so that this would not prove a critical constraint.
A small airline may not wish to undertake all its own maintenance, at least from the start. It will either have to rely on existing airlines (which may be impractical) or go elsewhere. Maintenance can be done on contract in other countries, such as Singapore or New Zealand. In the longer term, specialist maintenance operations may develop in Australia.
Airport and terminal facilities present perhaps the most difficult problem. It is often said that the Two Airline Policy has been set in concrete. Terminals in the major airports are often leased by Ansett and Australian, and it would be difficult to expand terminal space in a short time. This has not prevented new airlines from entering markets when they are permitted to, even when they are forced to resort to kerbside baggage handling. Such temporary arrangements enable a new airline to compete, even though it remains at a disadvantage compared to its competitors. It is significant that when airports like Adelaide became international gateways, it was possible to provide international terminals relatively quickly -- it should be easier to provide domestic terminals which require many less facilities.
The problem is one of reaching a situation whereby new entrants are not constrained from operating by unavailability of terminal space. Overseas experience in airlines (79) and coaches (80) confirms that preferential access to terminals gives a firm a strong market position. In Australia, there are two issues: first, the quantity of terminal space, and second, the terms of access. Where there is a shortage of space, it will take time to construct more. It should be remembered that, given the terms of the current Two Airline Agreement, change to the policy will take several years to achieve. The access problem is one that must be addressed immediately, however, lest decisions be taken now which limit possibilities in the future. The current situation, along with some of the possibilities, is well summarised by the May Committee. (81)
The government owns terminals and leases some of them to itself through Australian Airlines. If need be, it can do what it wishes with these terminals. While it is normally efficient to operate public enterprises at arm's length from the government, it could require Australian Airlines to sub-lease space to other entrants. This would put Australian at a competitive disadvantage vis-à-vis Ansett, but it could be compensated for this. A more extreme example would be to separate airline and terminal operation within Australian, and vest the terminal operation in the new Airports Corporation which could allocate the capacity to the highest bidder. This also would worsen Australian's position -- something which can be rectified through compensation. One of the possible advantages of public enterprise is that it can be used as an instrument of government policy.
Selling off terminals would limit the options open to the government, and there are unlikely to be any gains in operating efficiency through selling. Terminals can be owned by the government yet operated by lessees -- something which happens already to a degree. Privatisation simply means that the government gets more cash now, and less cash later. Indeed, it only gains cash to the extent that it sells to private operators -- a sale of some terminals to Australian Airlines generates no more cash for the government as a whole. If the government is obtaining a market return on leasing its terminals, it will neither gain nor lose in the long term from selling terminals.
The terminals could be sold to the main users, Ansett and Australian, or could be sold to third parties. From the point of view of operational efficiency, it is probably better that they be sold to the users, but from the point of view of access for new competition, selling to the users would be unsatisfactory. It would be possible, at the time of sale, to stipulate that the terminals be open to use by other airlines, but no firm likes being required to provide services for its competitors, and most firms will try to make life difficult for them.
The most satisfactory solution is for the government to retain ownership of the terminals, franchise their operation out, and guarantee access to all users, subject to availability of space. If franchisees fail to meet access guarantees, they can be replaced. Such a solution will be as remunerative for the government in the long term as selling terminals, but it will give it a stronger hand in imposing a pro-competitive policy. The major airlines can be given the opportunity to lease and operate the terminals they use, and gain operational benefits from this, as long as they conform to government policy on access. Once the government sells terminals, it is harder for it to influence the access problem in the future.
Over time, there may be shortages of airport capacity for aircraft movements. For example, at times Sydney Airport demand is pressing against capacity. This capacity is currently not rationed efficiently and many of the users pay virtually nothing for it. If capacity were rationed efficiently, for example by price, there would be adequate capacity to handle a considerable increase in demand by airlines. (82) Any overall problem of airport capacity need be no constraint on entry by new airlines.
PARTICIPATION BY OVERSEAS AIRLINES
An issue which needs to be resolved is that of the participation on domestic routes by the international airline, QANTAS, and by airlines from other countries. In other industries there has been no problem in allowing overseas firms to enter the Australian industry -- a good example of this has been the entry of foreign banks to Australia. From the viewpoint of economic efficiency, and benefits to Australia overall, it would be desirable to allow QANTAS and overseas airlines to operate on domestic routes. This would be especially true if it was considered that new firms might take some time to get established in Australia. In some cases international airlines, such as QANTAS and British Airways, already fly over domestic sectors; it is obviously wasteful for them to have capacity which they are not permitted to sell, as is the case at present.
From Australia's point of view, there would be no problems in allowing QANTAS access to domestic routes. Indeed, Australia is one of very few countries which restricts its international airline from doing this. Not only does it result in waste of capacity but it also hinders the airline in competing in overseas markets, through restricting what it can offer within Australia. There are some problems which do need to be resolved. To compete on equal terms, the international airline would have to pay the same rates for fuel and aviation facilities as the domestic airline. In addition, there would need to be some assurance that the airline would not cross-subsidise domestic routes out of profits earned on regulated international routes. (It ought to be noted that the owners of Ansett also have access to profits from regulated industries which they could, but do not, use to cross-subsidise the airline.)
Allowing international airlines from overseas access to domestic markets may appear to be granting them something for nothing. It is not -- it would be a matter of securing services provided in Australia from the most appropriate supplier. Australia would gain from this regardless of whether other countries granted access to Australian airlines. At present New Zealand allows Ansett's New Zealand joint venture airline to fly domestically even though Australia does not allow Air New Zealand to do the same. If reciprocity is to be the rule, it would be appropriate to allow Air New Zealand to operate within Australia. More generally, Australia can gain by allowing overseas airlines access. This right would be valuable to both sides, though it may be worthwhile to withhold such rights in order to negotiate more access for Australian airlines, e.g., QANTAS, in overseas markets (for example, less restrictions in the US market). Thus, it would be appropriate to decide in principle that access to overseas airlines be granted, even if rights were withheld until the countries negotiated a settlement satisfactory to Australia (not necessarily only aviation rights).
THE OWNERSHIP OF AUSTRALIAN AIRLINES
One of the two major airlines, Australian Airlines, is government-owned; a future government could wish to sell it. Here is not the place to consider the arguments for or against privatisation -- what is relevant here is whether its ownership would influence the working of deregulation.
If Australian Airlines were privatised, there should be no difficulties in its operating in a deregulated environment, along with other airlines. It is better that privatisation take place either after deregulation or simultaneously with it. This would avoid the clash of interests that governments inevitably find themselves facing when privatising a regulated firm. They may wish to improve performance by encouraging more competition but if they promise further regulation, they increase the price they receive for their firm. Thus, in the UK, the government has been unwilling to proceed with much airline liberalisation while selling British Airways. (83)
If Australian Airlines were to remain in government ownership, there would be doubts as to whether it would be a fair competitor. It could have access to subsidies which would enable it to drive more efficient airlines from the market. There are some grounds for these doubts -- in the early days, TAA operated at a loss and nearly drove the private airline bankrupt. Since 1950, no government has sought to use TAA/Australian in this way against Ansett, nor is it the policy of any major political party to do so in the future. There are several examples of government firms which compete with private firms where they are required to perform at or over market rates of return. The government's ownership of the Commonwealth Bank and the AIDC was not regarded as any hindrance to financial deregulation. The matter depends on the willingness of the government to ensure that its own firms act as fair competitors. If it wishes to drive private firms from the market place, it can use its own firm to do so, but it has many other ways of achieving the same result. Thus, successful deregulation is consistent with Australian Airlines remaining government-owned; to suggest otherwise is to raise either the ghost of long-past policies or a red herring.
RESIDUAL REGULATION
It is possible to achieve economic deregulation of an industry while retaining other forms of regulation. Thus there may remain regulation of the labour force through the arbitration system, while the industry itself is deregulated. An industry may be subject to general legislation applicable to all or most industries, such as that to control restrictive trade practices. Some of the doubts concerning anti-competitive practices, such as predatory pricing and mergers, under deregulation can be addressed in the same way as for other industries.
Safety regulation is likely to be something which is considered at the time of deregulation. In the US, safety regulation was retained, and to a degree, strengthened. There is likely to be no perceptible effect of deregulation, partial or full, on the safety of operation of airlines. US experience strongly indicates this -- the safety performance of US airlines continued to improve over time, and was not perceptibly affected by deregulation. (84) If any changes in economic regulation had any adverse effects, these could be countered by tighter safety regulations if these were considered desirable. Like all forms of regulation, safety regulation is not always effective, and sometimes backfires. It may be questioned whether every instance of it is necessary or desirable. (85) Whatever the form of economic regulation, an appropriate form and level of safety regulation can be chosen to achieve safety objectives.
10. MANAGING CHANGE: THE TRANSITION TO DEREGULATION
Suppose that the government has decided that it would like to deregulate, wholly or partly, the airline industry. It then must decide how to achieve this. If there is widespread agreement on its desirability, and no-one is adversely affected, it might as well do it immediately. However, with airline deregulation, this is not going to be the case -- some groups may lose from it, and they will oppose it. These groups may have political or industrial muscle, and they may be able to force the government's hand (pilots, for example, are credited with forcing the government to alter its proposal for taxation of superannuation benefits). Two key issues emerge in determining the path to deregulation -- one is the speed of adjustment, and the second is the possibility and desirability of compensation for adversely affected groups.
THE SPEED OF ADJUSTMENT
This is not an analytical so much as a political question. If a move is proposed which involves an improvement from an overall national point of view, it is efficient to make the move immediately: it is best to obtain the gains for as long a period as is possible. Various parties may take time to adjust, and they may lose from the change, but delaying the change holds up benefits which are greater than the costs. Parties would know the full extent of change required, and they can choose to adjust at their own speed. Only if it is known that individuals and firms would make systematic and predictable mistakes (e.g., too many firms would enter) might a slower change be more efficient.
Immediate change may not be the best or most practical way when some groups lose out. Thus, when introducing major changes, governments often phase them over considerable periods; the Australian car plan is an example. Moving faster may be regarded as politically impossible.
It should also be noted that an immediate deregulation is not possible under current legislation. If the Two Airline Agreement is to be terminated, three years' notice must be given. It may be possible for a government intending to deregulate to use its discretion in a more liberal way from the time of giving notice, but major changes, such as entry on to trunk routes, are unlikely to be possible until the agreement has been terminated. Within that period, preparations can be made -- terminals can be built, aircraft purchased, and potential entrants and incumbents can devise their strategies. Also, within this period, the government can devise the transitional arrangements most appropriate for the environment it is seeking to achieve. It would be possible to have gradual or immediate deregulation on termination of the agreement.
There are examples of the government implementing major changes quite rapidly. One is the US airline deregulation. (86) Another is the deregulation of Australian financial markets. (87) In both of these cases, no groups felt they were likely to be major losers, though many had doubts concerning how changes would affect them. Changes were not all instantaneous, but they followed one another rapidly, and the timing of particular changes was sometimes advanced. The staged, but rapid, changes had the advantage of reassuring parties that they could survive and prosper under the new arrangements, made further changes a logical consequence of previous moves, and gained approval for them. The process of change took on a momentum of its own. If all that needs to be addressed are doubts about how a new system will operate, this will be the best way of proceeding.
This is not the case with Australian domestic airlines -- while there are doubts about how deregulation would operate, some groups would see themselves as likely losers and would be prepared to fight any change. A rapid deregulation would be possible, but only if the government was prepared to take on opposing groups and was able to override them. The way the Two Airline Policy has survived with little alteration since the late 1950s is evidence that previous governments have not been prepared to make changes in the face of opposition (especially in 1981). The opposition by the airlines to change is now perhaps less than before, while that by the unions, mindful of US experiences, is stronger.
Granted this, would gradual change be possible? One option is to try to identify changes that would produce benefits but not impose substantial losses on any groups. It would be possible to implement these fairly readily. Such changes would produce less overall benefit however. They might be regarded as a first step towards more thorough change -- if the latter is to be attempted, the government must eventually address the problem of opposition from loser groups. It is conceivable that successful implementation of limited change may strengthen the support for total deregulation though the reverse could also happen -- it would take the urgency out of the issue. The best example of such a half-way house would be the introduction of a competitive charter sector. Whatever the pattern of change, it is naturally desirable that those gains which are costless be had.
Suppose that the government chose to implement gradual change towards full deregulation. Two points can be made about this process. First, if change is gradual, it may be possible for groups which oppose it to halt the process in its tracks. If total deregulation is desired, it may be preferable to do it rapidly, before opposition crystallises. A government may find it difficult to convince others of its seriousness in deregulating gradually. Governments have been committed to full cost recovery for airports and air navigation facilities for over a decade, yet they still have not achieved it for general aviation. In such a case, it will be desirable for government to constrain itself by undertaking some commitment which is difficult to break (e.g. by entering contracts with firms allowing them to establish airlines by particular dates). The Two Airline Policy itself is an excellent example of the government making its long-term commitment to a course of action credible by entering contracts with other parties.
A second point is that it may be possible to minimise the adverse consequences on particular groups by moving gradually or at least by managing change in such a way that it consists of a series of barely perceptible alterations never worth opposing. Personal tax rates have increased substantially in the past 15 years -- this has happened gradually, and it is difficult to believe that the same tax rise could have been implemented at one point of time. Groups which lose out may not be sufficiently motivated to oppose gradual change, even though they would certainly oppose the same change if it happened over a short period. With gradual change it may be possible to guarantee the position of those currently in the system, while applying different rules for those who enter it later. This is the method used for altering arrangements for taxation of superannuation benefits. Those who lose are only losing out as compared to what they might have had, and since they are dispersed and not in the system, they cannot mount effective opposition. Slow implementation of policies which are unpopular with some groups is often attempted. In the case of airline deregulation, the difficulty lies in designing changes which are gradual in their impact: often, changes must be of an all-or-nothing variety. One form of policy would be one of gradually lessening restrictions on capacity.
ACHIEVING CHANGE THROUGH COMPENSATION
If there are groups which will lose through deregulation, and they have sufficient power to halt change or make it costly for the government to proceed, another option is for the government to compensate them. When the US deregulated airlines, there was concern about the effect on the small communities and about possible job losses. These fears were allayed by commitments to subsidise local routes if need be, and compensation for employees who lost jobs because of deregulation. This enabled deregulation to go ahead. In fact, some other groups did lose, and the system of compensation for job losses proved unworkable, but these problems emerged after the steps had been taken. The same approach could be taken in Australia.
Several aspects of the compensation approach need to be addressed. First, compensation is rarely costless, even in overall efficiency terms. It may involve charges which mean that 90 per cent or 10 per cent of the possible potential benefits are gained rather than none of them. If, for example, low-density routes are subsidised, there will be an efficiency loss in that some people obtain services they would not be prepared to pay for. Often a scheme to achieve one set of changes will put in place a new set of inefficiencies -- when ports were containerised, the labour force was reduced, but this was only made possible by giving a strong monopoly to unions and those already in the industry. The inefficiencies of the labour system in ports is a long-standing and intractable problem (it too will probably have to be solved with use of compensatory mechanisms).
Second, it must be understood that compensation is being paid not because it is equitable to do so, but because it is the price of securing agreement to change. If a move to deregulation disadvantaged pilots, it might be appropriate to compensate them, not because they are especially deserving, but because they may be able to frustrate change.
Third, it may be argued that granting compensation will be paying off those who have used monopoly power to raise their own incomes, and that this will give other groups the incentive to create monopoly power to be bought off later. If it is possible to create the monopoly, there will be the incentive to do so.
In Section 6, several groups which may lose have been identified. Some, such as coach operators, would be difficult to compensate, but they would be unlikely to lose much. Others, such as rural communities which may lose air services, can more easily be compensated. As argued above, it is improbable that these would be much affected by deregulation, but it may be worthwhile devising schemes to guarantee their services -- they will often not be necessary, and they will be cheap if they are. Passengers on particular flights (say medium-density flights) may pay higher fares, but they will probably also gain when they travel on high-density routes. The two major potential losers are the major airlines and their employees.
One of the airlines is owned by the government, so that any adverse effects faced as a result of deregulation will be ultimately faced by the government -- if it seriously wishes to deregulate, it will be willing to absorb them. As indicated, the private airline, Ansett, may be worse off. It would be difficult to arrange matters so that it enjoyed the same level and stability of profit as under the Two Airline Policy. So too it is difficult to put any monetary amount on its losses because these would depend on how the Policy was operated, and it is quite possible that it will do better under deregulation than if its profits were tightly regulated under the Policy. There are ways in which the government could compensate -- one of the more likely is to grant the airline access to more international routes. In principle, compensation in cash might be possible, through it would be politically risky.
There is uncertainty about what will happen under deregulation, and perhaps the best strategy is to try to address the main concerns that worry the airline. First, it may be worried that QANTAS, if allowed to compete domestically, will have access to government funds and to profits on regulated routes, with which to cross-subsidise domestic routes and drive the private domestic airlines out. This may or may not be a legitimate worry, but it may be best handled by restricting QANTAS's access to domestic routes. It is efficient to allow QANTAS sufficient rights to fill aircraft on domestic segments of international flights; there is waste of capacity at the moment. However, there may not be any particular advantage, through increasing competition, in allowing QANTAS to compete more extensively on domestic markets; there may be sufficient competition from purely domestic airlines. An efficient QANTAS may not be able to supply a particular service at a price any lower than an efficient domestic airline. While it is desirable to use QANTAS's excess capacity, it may not be necessary to have QANTAS as a full competitor, and if there are doubts concerning its ability to cross-subsidise, then it may be best to segregate domestic and international traffic. The main cost of this would not be through lessened competition, but through weakening QANTAS's marketing position overseas.
The second fear of Ansett is one of competing against a government-owned airline -- this is a long-standing, but probably outdated, concern. It is significant that Ansett is investing heavily to set up an airline in New Zealand to compete against a (bigger) government airline in a deregulated market -- precisely the situation it states is unacceptable in Australia. The owners of Ansett compete in other industries against government firms, some of which, such as railways, are subsidised.
Allaying the fears of the labour force may be the most difficult task. In a freer environment, many employees' positions may be eroded, and their remuneration will fall, though most likely they will preserve their jobs. It is possible to give guarantees of jobs, and to compensate anyone who loses a job through deregulation, but this would be very complex to administer. There can be agreement to maintain the present industrial relations structure, with award wages and conditions to be preserved. Notwithstanding this, employees may be concerned that in a deregulated environment there might be an erosion of award wages and conditions. This would seem quite likely to those who consider that key determinants of wages and conditions are demand and supply factors, which depend on the competitiveness of the industry; wages and conditions depend on the environment, and not just the industrial relations machinery in place. For example, unionisation is very high in the industry at present -- under deregulation, this situation may be difficult to maintain. While unions profess to be satisfied that this machinery will guarantee their position, (88) their opposition to deregulation suggests that they are not entirely convinced by their public position.
Continued regulation of the airline labour market is consistent with the deregulation of the airline product market, and it will go some way towards preserving the position of employees. This regulation may be questioned, however. It can be argued that working conditions are contributing to the low labour productivity of airlines. If they are preserved, opportunities for improved efficiency under deregulation will be missed. Even if wage levels are maintained, flexibility in work practices is necessary to enable benefits from improved efficiency to be achieved. One might also object that the aviation labour market is not an efficient or equitable one (why should pilots be paid so much more than air traffic controllers?), and that there is a case for deregulating it. This could result in large changes in remuneration, which would be resisted. The merits or demerits of this are not discussed here.
If it is a matter that current employees may lose from deregulation, and they will oppose change, it may be possible to preserve the position of current employees while introducing longer-term changes. Thus, current employees may be given "grandfather rights" to a job at current pay, while new employees may be paid less. This has been happening in the US with established airlines; it has its problems, in that older airlines have higher cost structures than newer airlines, and this in turn has led to problems at the product market level. The smaller the changes in the labour market, the smoother will be the deregulation process, though its effects will not be as thorough.
The group most affected by deregulation is also the group most difficult to compensate. Labour market regulation can limit the adverse effects on some individuals, but it can also limit the gains from improved productivity. Also, it may not prove very effective in a deregulated environment. If, as is possible, the losses to some individuals are large, it may be much easier to deregulate gradually, rather than at once. It may be possible to move to a half-way house, with a competitive charter industry, under which no group loses significantly, and address the problem of compensation over a larger time frame.
Many cases of deregulation occur when everyone believes they will gain or at least not lose from it. Examples include US airline deregulation and Australian financial deregulation. Such cases lead to straightforward changes which tend to gather their own momentum. At other times, specific groups will lose, or at least fear they will lose, from a change in regulation; they will hence oppose it. If they are strong, it will be difficult for the government to deregulate. The way the Two Airline Policy has survived despite attempts to change it reflects the benefits it produces for specific groups. Deregulating the airlines in Australia will be a difficult process. Either the government must be prepared to override powerful groups or devise means for compensating them in ways that do not negate the benefits of deregulation.
11. CONCLUSIONS
In this paper, the focus has been on the problems of moving to a deregulated airline environment, and on what such an environment might be like in terms of industry structure and the competitiveness of firms. Such a focus -- on problems rather than advantages -- may seem negative. However, there is now a good chance Australia will move in the direction of a more deregulated airline system and it is important to know just what this might involve. With any major change, there are bound to be costs of adjustment, and the better they are appreciated beforehand the lower they will be.
It is improbable that a deregulated airline industry will be a perfectly competitive one. Few, if any, industries are. Nor will it be perfectly contestable -- potential competition will have an effect, but it will not be strong enough to force perfectly competitive behaviour on the firms in the industry. In this respect, the airline industry will be much like many other unregulated industries, such as retail, road freight, air freight, inter-state coaches, and restaurants. It will be possible to point to market imperfections and cases of market power, and these will lead to performance short of the ideal. Regulation is not often suggested here as a means of eliminating these imperfections, because it will create its own, larger, problems. The problems with the current regulatory system, the Two Airline Policy, are major, and a deregulated environment can be expected to result in a clear gain in terms of economic efficiency.
Any inefficiencies in such an environment are likely to be minor compared to those which are experienced now.
A number of specific issues have been considered here. The US airline deregulation has been a model for what might happen elsewhere. Given that the starting point, in terms of the nature of regulation, is different in Australia, different changes can be expected to take place. Changes in service patterns and networks are likely to be less extensive. The structure of air fares may not change as much, but there will be more extensive use of lower, but restricted, fares. There is more scope for improvements in productive efficiency within the airlines, and new entrants will force the existing airlines to take up this scope.
The airline industry is going to be neither a perfectly competitive nor contestable one. The degree of actual competition will have an influence on results. Thus, there is likely to be sufficient competition to force prices to cost on the busy routes on which about two thirds of travellers fly. On some but not all other routes, active or potential competition will not be strong enough to achieve the same results. Airlines serving them will have some market power, and they will try to use it. Overall, there will be efficiency gains and fare reductions, though this may not apply on every route.
There is scope, in a deregulated environment, for a few large airlines with nationwide networks, like Australian and Ansett. It is possible that one may dominate, and if this happens, it is likely, at this stage, that Ansett will be the dominant airline. Currently, with its regional subsidiaries, it has a wide network, it is more profitable, and it has less restrictions placed on its commercial freedom by the government. A realisation of this may underlie Ansett's more positive approach to deregulation. While Australian may stay in government ownership, it will need to be given more freedom than at present for it to prosper.
A situation with a large, dominant, airline with perhaps 50 per cent of the market is not ideal, but it would be one with less monopoly power than at present. The large airline would be held in check by smaller airlines, one or two of which may be large enough to present a threat to its dominance. The situation might be compared to that of the retail supermarket industry, with Coles-Myers being the dominant supplier. The competition that exists is sufficient to prevent Coles-Myers from acting as a monopolist. The barriers to entry into the airline industry will remain small.
A major change in industrial structure rarely comes with no one losing. In this case, some specific sections of the labour force, those which have done well out of regulation, will lose, though the labour force as a whole will gain, with more jobs in the industry. The market power of some groups of workers who are specific to the industry will be weakened -- the extent to which they lose depends on the effectiveness of unionisation and the degree of labour market regulation. Regulation may limit the effects of changes taking place in the product market, but it is unlikely to be strong enough to preserve present positions entirely. In turn, the form of labour market regulation affects how airline deregulation works. If regulation results in little change in working conditions or pay, it will enhance the position of existing firms. If it does not, the position of new entrant airlines will be helped by the fact that they are not constrained to old agreements, and they may enjoy cost advantages over the established airlines for several years.
There are several alternatives to complete deregulation, but one which has been given little attention is that of competitive charters. This would involve allowing the development of a charter airline segment of the market, which would be open to entry but which would be limited in ways so as to prevent it from direct competition with the scheduled airlines. Its main advantage would be that it would develop and serve the low-fare, leisure/tourist, market, which is relatively neglected by the scheduled airlines. Its disadvantage is that it would put little pressure for efficiency improvements on them. If the objective is to achieve some benefits with little disruption to the established firms and their labour forces, this could be regarded as an advantage. It could be regarded as a stage in a move to complete deregulation.
Managing the transition to a deregulated industry is a delicate problem. If all affected groups are in favour of deregulation, then there should be no problem in achieving it as soon as current legislation and agreements allow. This is not likely to be the case however: when groups perceive, rightly or wrongly, that they will lose, it is difficult to achieve change, especially in Australia.
One option is to move slowly, so that the losses experienced at any point of time are small, and change is not worth opposing. The difficulty is that this gives opportunities for opponents to halt progress.
Another option is to guarantee the position of those already in the industry, while accepting others into the industry on less advantageous terms. This has been done in the US with some airlines' labour forces, but it can lead to instability (new firms have an advantage over old firms even if they are no more efficient).
Finally, support for change can be obtained from potential losers if compensation is offered. This was done in the US -- successfully with small communities, and unsuccessfully with labour. If deregulation is seen as likely to worsen service standards to rural communities, then subsidies to maintain standards will be a minor cost in comparison to the benefits elsewhere. Compensation of those in the labour force who lose will be more difficult to achieve. If it is done through tighter labour force regulation, there will be the costs of this regulation to consider. For example, it may limit gains from deregulation by restricting the flexibility with which work practices can be changed. There is a long-term problem with compensation as well: if those who have enjoyed profits from monopoly situations are compensated when monopoly is taken away, the incentives for creation of other monopolies will be greater.
Swift change, without compensation, is probably the most efficient, but it can be seen as inequitable, and those who lose will fight to block it. There are many examples of regulation in Australia which have been recognised for many years as promoting inefficiency, but which have survived, with little alteration; one is regulation of the dairy industry, which has survived because governments have not been willing to impose losses on small, but identifiable, groups. The durability of the Two Airline Policy itself is due, in part, to this reluctance. The most difficult problem in changing regulation of airlines in Australia is not in choosing what option is most acceptable, but in devising means of implementing the chosen option.
ENDNOTES
41. T.E. May (chairman), Independent Review of Economic Regulation of Domestic Aviation (May Report), AGPS, Canberra, 1986.
42. K.R. Mackay, "A comparison of the relative efficiency of Australian domestic airlines and foreign airlines" in Department of Transport, Domestic Air Transport Policy Review (Vol. 2), AGPS, Canberra, 1979; M.G. Kirby, Airline economies of scale and Australian domestic air transport policy, Discussion Paper No 112, Centre for Economic Policy Research, ANU, Canberra, 1984.
43. P.J. Forsyth and R.D. Hocking, Economic Efficiency and the Regulation of Australian Transport, CEDA M Series No. 62A, Council for the Economic Development of Australia, Melbourne 1980; Bureau of Transport Economics, Competition and Regulation in Domestic Aviation: Submission to Independent Review, Occasional Paper No. 72, 1985; and May Report, Appendix 5.
44. P.J. Forsyth, Can international comparisons of air fares be made? Discussion Paper No. 133, Centre for Economic Policy Research, ANU, Canberra, 1985.
45. R.S. Doganis, Flying Off Course: The Economics of International Airlines, Allen and Unwin, London, 1985; E.E. Bailey, D.R. Graham, and D.P. Kaplan, Deregulating the Airlines, MIT Press, Cambridge, 1985.
46. May Report, Chapter 20.
47. For example, see P.J. Forsyth, US airline deregulation: An interim assessment, Institute for Fiscal Studies Working Paper No. 29, July 1981; J.R. Meyer and C.V. Oster, Deregulation and the New Airline Entrepreneurs, MIT Press, Cambridge, 1984; Bailey, Graham and Kaplan, op. cit.; N.K.Taneja, "US Airline Deregulation", and J.R. Meyer, "The American Experience with Airline Deregulation", papers presented at Seminar on Economic Regulation of Aviation in Australia, Bureau of Transport Economics, AGPS, Canberra, 1985; S. Morrison, and C. Winston, "The Economic Effects of Airline Deregulation" in Studies in the Regulation of Economic Activity, The Brookings Institution, Washington, 1985; Bureau of Transport Economics, Deregulation of the Domestic Airline Industry in the United States, Information Paper No. 16, AGPS, Canberra, 1986.
48. For example, see G.W. Douglas, and J.C. Miller, Economic Regulation of Domestic Air Transport: Theory and Practice, Brookings Institution, Washington, 1974.
49. For example, see D.W. Koran, "The Welfare Effects of Airline Fare Deregulation in the United States", Journal of Transport Economics and Policy, Vol. 17, 1983; and most recently, Morrison and Winston, op. cit.
50. Bailey, Graham and Kaplan, op. cit. Chapter 3.
51. Morrison and Winston, op. cit.
52. Meyer, "The American experience with airline deregulation", op. cit.
53. For example, see K. Page, "FAA increases safety scrutiny", Airline Business, May 1986.
54. On contestability, see W.J. Baumol, "Contestable markets: an uprising in the theory of industry structure", American Economic Review, March 1982.
55. E.E. Bailey and J.C. Panzar, "The contestability of airline markets during the transition to deregulation", Law and Contemporary Problems, Winter, 1981; though see also D.R. Graham, D.P. Kaplan and D.S. Sibley, "Efficiency and competition in the airline industry" Bell Journal of Economics, Spring 1983, and D. Starkie and M. Starrs, "Contestability and sustainability in regional airline markets", Economic Record, Vol. 60 No. 170, September 1984, pages 274-283.
56. G.D. Call and T.E. Keeler, "Airline deregulation, fares and market behaviour: some empirical evidence" in A.F. Daughety, Analytical Studies in Transport Economics, Cambridge University Press, Cambridge, 1985; and E.E. Bailey and J.R. Williams, "Sources of economic rent in the deregulated airline industry", paper for Graduate School of Industrial Administration, Carnegie Mellon University, April 1986.
57. Douglas and Miller, op. cit.
58. Bureau of Transport Economics, Australian Long-Distance Coach Industry Review, Occasional Paper No. 74, AGPS, Canberra, 1985.
59. Ibid., pp. 101-104.
60. Baumol, op. cit.
61. W.A. Brock and D.S. Evans, "Creamskimming", Chapter 4 of D.S. Evans (ed.), Breaking Up Bell: Essays on Industrial Reorganisation and Regulations, Elsevier, New York and Amsterdam, 1983, p. 71.
62. Bailey and Williams, op. cit.
63. Graham, Kaplan and Sibley, op. cit.; Call and Keeler, op. cit.
64. For a discussion of responses, see Call and Keeler, op. cit.
65. Bailey and Williams, op. cit.
66. Bureau of Transport Economics, Australian Long-Distance Coach Industry Review, op. cit., p. 26.
67. Department of Aviation, Domestic Air Transport Statistics 1983-84.
68. Bailey, Graham and Kaplan, op. cit., p. 62.
69. May Report.
70. For example, see P. Milgrom and P. Roberts, "Predation, reputation and entry deterrents", Journal of Economic Theory, Vol. 27, 1982, pp. 280-312.
71. May Report.
72. For example, see Treasury, Competition in the Air, (Treasury Submission to Independent Review of Economic Regulation of Domestic Aviation), Treasury Economic Paper No. 11, AGPS, Canberra, 1985.
73. R.P. Albon and M.G. Kirby, "Cost-padding in profit-regulated firms", Economic Record, Vol. 59, 1983, p. 164.
74. The British regulations are discussed by R. Baldwin in Regulating the Airlines: administrative justice and agency discretion, Clevedon Press, Oxford, 1985.
75. M.G. Kirby, Airline Economics of Scale and Australian Domestic Air Transport Policy, Discussion Paper No. 112, Centre for Economic Policy Research, ANU, Canberra, 1981, and C. Findlay, "Occupational Regulation in a Regulated Industry: The Case of Airline Pilots" in Albon and Lindsay (eds), Occupational Regulation and the Public Interest, Centre for Independent Studies, Sydney, 1984.
76. Findlay, op. cit.
77. For some discussion of airline labour market issues, sec I. Oldmeadow, "Industrial Perspectives on Regulation", paper presented at Seminar on Economic Regulation of Aviation in Australia, Bureau of Transport Economics, AGPS, Canberra, 1985.
78. Morrison and Winston, op. cit.
79. Bailey and Williams, op. cit.
80. S. Jaffer and D. Thompson, "Deregulating Express Coaches: A Re-assessment", Fiscal Studies, Vol. 7 No. 4, November 1981, pp. 45-68.
81. May Report, Chapter 28.
82. G. Mills, Investment in Airline Capacity -- A Critical Review of the MANS (Major Airport Needs of Sydney), Study for Australian Transport Research Forum, Hobart, 1982; and P.J. Forsyth, "Airport Runway Capacity in the Sydney Region" in Changes in the Air? Issues in Domestic Aviation Policy, Centre for Independent Studies, Sydney, 1984.
83. M. Ashworth and P.J. Forsyth, Civil Aviation Policy and the Privatisation of British Airways, Institute for Fiscal Studies, Report Series No. 12, London, 1984.
84. See section 7 of Osier's piece in this volume; also Page, op. cit.
85. M.G. Kirby, Domestic Airline Regulation: the Australian Debate, Centre for Independent Studies, Sydney, 1981, Chapter 4.
86. A.E. Kahn, "Is it time to re-regulate the airline industry?" The World Economy, Vol. 5 No. 4, 1982.
87. I.R. Harper, Why Financial Deregulation? Discussion Paper No. 132, Centre for Economic Policy Research, Australian National University, October, 1986.
88. Oldmeadow, op. cit.
No comments:
Post a Comment