The Impact of Discount Airlines on Domestic Fares in Canada.

The oligopolistic tendencies of the airline industry are even more pronounced in Canada than in the United States. In the late 1990s, only two airlines, Air Canada and Canadian Airlines, accounted for approximately 80 percent of industry operating revenues, according to Transport Canada, which

is the government ministry responsible for transportation. [1] In the face of such extreme market concentration, one would expect low-fare carriers such as WestJet and Canada 3000 to be especially important in suppressing fares of the major airlines, similar to the impact of Southwest Airlines in the United States. However, the impact of low-fare carriers in the Canadian context has not been tested empirically.

The purpose of this article is to determine whether the presence of a low-fare carrier on a domestic Canadian route suppresses the fares of the major airlines. First, background material on the Canadian airline industry is provided. Next, regression analysis is applied to data on 163 city-pairs within Canada to identify the effect, if any, of low-fare carriers on major airlines' fares. Last, the implications for public policy are discussed briefly.

THE NATURE OF THE AIRLINE INDUSTRY AND MARKETS IN CANADA

Traditionally, Air Canada and Canadian Airlines have dominated the industry. CP Air and its successor, Canadian Airlines, were stronger in the western half of the country and in the transpacific market, while Air Canada was stronger in the eastern half of the country as well as the transatlantic and Caribbean markets. Deregulation of domestic routes began with the National Transportation Act of 1987; thereafter, Canadian Airlines increased service in the eastern half of Canada, where it had traditionally been weak. International deregulation began in 1994, when the Minister of Transport terminated what had been known as the "Division of the World" policy. This led to Air Canada entering the transpacific market and Canadian Airlines increasing transatlantic services. (In the airline industry in Canada, the term "international" typically refers to flights to countries other than the U.S., and that usage is followed here.) Deregulation of transborder flights (i.e., Canada-U.S.) began in 1995 with the Open Skies Agreement. [2]

The process of deregulation following the National Transportation Act of 1987 led to two carriers, Air Canada and Canadian Airlines, in constant head-to-head competition on every significant domestic route, as well as the more important transborder and international routes. The two airlines typically offer identical fares, and when one airline announces a "seat sale," the other matches those fares within a day. In terms of financial results, Canadian Airlines had chronically been the weaker and smaller of the two airlines. Air Canada's operating revenues have been far greater than those of Canadian Airlines: C$5.9 billion for Air Canada versus C$3.2 billion for Canadian Airlines for 1998. For the period 1993-98, Canadian Airlines reported a loss in five of the six years, showing a profit only in 1997. During the same period, Air Canada reported losses in two of those six years, 1993 and 1998. [3]

Because the airline industry offers a relatively pure model of market competition, it has long been the focus of study by researchers in the United States. The costs of adding or deleting a city from one's route network are minimal when compared to the barriers of entry, for example, in building a factory or a store. Nevertheless, at some U.S. airports a dominant airline controls nearly all gates, making it difficult for another airline to begin service. The term "fortress hub" has been coined to describe a situation where an airline has a degree of dominance that is very difficult to overcome. The hub concept is less relevant to Canada, where the major cities form a linear pattern across the southern part of the country. Nevertheless, some airlines in Canada have attempted to use the hub concept; for instance, Canada's most important hub, Toronto, has always been dominated by Air Canada.

In addition to the issue of fortress hubs, another phenomenon impeding market forces in the airline industry is the "golden rule." [4] This occurs when an airline is reluctant to reduce fares on a given route against a competitor, for fear that the other airline will retaliate by reducing fares elsewhere on routes served by both airlines. In other words, airlines are reluctant to initiate a price war on one route because their competitor will retaliate on a route where that competitor is particularly strong. The "golden rule" applies mainly to large competing airlines with extensive, overlapping route networks. Although the term originated in the U.S., it may explain why there was so little competition in fare levels between Canada's two major carriers throughout the 1990s.

According to the U.S. Department of Transportation, the presence of a low-fare carrier serving a city-pair resulted in significantly lower fares in that market, and this effect was especially strong if one of the cities in the pair was an airline hub or if the route was less than 750 miles. [5] In such instances, the presence of a low-fare carrier reduced the average fare between the city-pair by more than half. The data showed that low-fare carriers seemed to be the only source of competitive pressure in short distance markets, defined as those under 750 miles. On these shorter routes, even the presence of a large number of airlines did not lead to lower fares unless one of the airlines was a low-fare carrier. The U.S. Department of Transportation defined low-fare carriers to include Airtran, Vanguard, and Southwest Airlines, of which Southwest was by far the largest.

Although the relevant research has focused almost exclusively on the U.S., there is every reason to expect similar effects in Canada. Throughout the 1990s, the Canadian domestic airline industry was characterized by the two major carriers, some small airlines, and a variety of low-fare carriers such as WestJet and Canada 3000, among others. One would expect the presence of a low-fare carrier to result in lower fares between city-pairs, similar to the effect that has been observed in the U.S. Hence, the following was proposed:

Hypothesis: The presence of a low-fare carrier on a route is associated with lower fares on the major carriers serving that city-pair.

RECENT DEVELOPMENTS IN THE AIRLINE INDUSTRY IN CANADA

After the data had been collected for this study, the Canadian airline industry went through a convulsive restructuring, culminating in the acquisition of Canadian Airlines by Air Canada. [6] The event that triggered the restructuring was Canadian Airlines' announcement in March 1999 that it was approaching insolvency. On August 24, 1999, Gerald Schwartz, CEO of the Toronto-based conglomerate Onex Corporation, announced his intention to acquire both Canadian Airlines and Air Canada, with financial support from American Airlines. The takeover bid had the complete cooperation of Canadian Airlines but was unwanted by Air Canada. Meanwhile, on October 19, 1999, Air Canada announced a hostile takeover bid against Canadian Airlines, thus creating a situation in which each of the two major carriers was attempting a hostile takeover of the other. On November 5, 1999, a court ruled that the Onex/Canadian Airlines bid to buy Air Canada was illegal, because American Airlines' role violated foreign ownership rules. This court decision cleared the path for Air Canada's hostile takeover of Canadian Airlines, which was completed on December 8, 1999. Air Canada has stated that it will continue to operate Canadian Airlines under its current name for the foreseeable future as a subsidiary of Air Canada.

As a result, 1999 ended with Air Canada controlling 80 percent of the domestic airline market. In February 2000, Canada's Minister of Transport, David Collenette, announced the reregulation of the airline industry. [7] The proposed legislation would increase the power of the Competition Bureau and the Transport Canada to control fares, and even has provisions for prison terms for executives and directors of offending airlines. Furthermore, Collenette stated that if domestic competition didn't develop within two years, U.S. airlines would be allowed to serve domestic routes in Canada (i.e., cabotage). Collenette's draconian reregulation policy was likely inspired by Air Canada's announcement of a fare increase in early December 1999 -- an incredible blunder of timing -- which exacerbated fears that Air Canada would exploit its near-monopoly position to engage in price-gouging.

Thus, the airline industry in Canada has come full circle: It began as a regulated industry, followed by deregulation, followed by a hostile takeover leading to a near-monopoly situation, which in turn necessitated reregulation. In short, Canada's experiment in airline deregulation has been a failure.

DATA AND VARIABLES

Data

Data on fares were obtained from the Microsoft Expedia website, [8] which offers access to a database of schedules and airfares, equivalent to what can be accessed by a travel agent. In accordance with industry custom, the Expedia database expresses all fares on a round-trip basis. Since airfares change constantly, November 11, 1998 was chosen arbitrarily as the point at which data were retrieved from the website for this study. That date had the advantage of being between seat sales; the seat sales that followed Air Canada's 1998 strike had come to an end, and the January 1999 seat sales had not yet been announced. On November 11, 1998, neither of the major carriers was promoting any unusual discounted fares on domestic routes. However, in the airline industry, there is no typical time of the year, and it is possible that different results would have been obtained if the study were conducted, for example, immediately before the peak summer season. The data were collected prior to Canadian Airlines' announce ment of impending insolvency in April 1999 and prior to the hostile takeover of Canadian Airlines by Air Canada in December 1999.

The study was limited to domestic Canadian airfares, rather than transborder or international flights. This limitation was imposed because transborder and international charter flights often consist of air-and-hotel packages, especially for Mexican and southern U.S. destinations, making it especially difficult to compare charter carriers' fares to those of the major scheduled carriers.

A list was compiled of every possible city-pair among the twenty-five most populous metropolitan areas in Canada as of 1996, as defined by Statistics Canada. However, five of those metropolitan areas had no scheduled air service, and for certain other city-pairs, the Expedia database did not show any airfares. The resulting data set consisted of 164 city-pairs. Merely because Expedia showed an airline offering a fare between two cities does not imply that the airline offered a direct flight on that route; for some city-pairs, an airline might require one or more connections. Also, commutativity was assumed; that is, the fare from Toronto to Montreal and return is the same as the airfare from Montreal to Toronto and return. This assumption of commutativity, which incidentally does not hold true for transborder or international flights, halved the size of the data set from 328 to 164.

Airlines in the Study

Until late 1999, the Canadian domestic airline industry was dominated by two major carriers, Air Canada and Canadian Airlines. [9] A second tier consists of low-fare carriers, such as Royal Airlines and Air Transat. These are charter operations, although the frequency of flights is so regular that they take on some of the characteristics of scheduled carriers. Also, there are a variety of airlines that could be described as minor scheduled carriers, such as Bearskin Airlines, First Air, Harbour Air, SaskAir, and others. These airlines usually offer low fares on routes where they compete against the major carriers, often in the form of foregoing advance-purchase and/or Saturday stayover requirements. An airline that constitutes a special case is Canada 3000, which had been a charter carrier but now appears in airfare data banks as a scheduled airline. Another fairly unique airline is WestJet, a low-fare scheduled airline not listed in the airfare data banks and operating outside the customary agents' commissi on structure. In its strategy, WestJet would be most similar to Southwest Airlines in the U.S. (The now-defunct Greyhound Airlines would have fallen into this category as well.)

For purposes of this study, the phrase "major carriers" refers to Air Canada and Canadian Airlines, while "low-fare carriers" refers to all other air carriers, whether charter or scheduled, serving domestic routes in Canada.

To assess which variables are associated with low airfares in various city-pairs, the key variable of this study was the number of low-fare carriers serving a given route. (Insofar as Air Canada and Canadian Airlines are concerned, in this study there were no city-pairs that were served by one but not the other, until the industry restructuring in early 2000.)

Dependent Variables

Multiple regression was used in this study. The dependent variable was the lowest fare offered by a major carrier (Air Canada or Canadian Airlines) between a given city-pair. Two versions of the dependent variable were used: Lowest fare requiring a Saturday stay-over, and lowest unrestricted fare. This was of interest due to the assumption in the airline industry that low-fare carriers draw off non-business travelers, for whom a Saturday-stay-over requirement is not an impediment, while business travelers will stay with the major carriers because they are not sensitive to price and not willing to extend a business trip into the weekend.

Independent Variables

Of the independent variables, some served as control variables. One such independent variable was the distance between the two cities in each city-pair, which is likely to be strongly related to the airfare. It was also necessary to control for the population of the cities in each city-pair, since larger cities would be expected to have higher traffic levels and therefore lower fares. [10] Population was expressed as that of the census metropolitan area as defined by Statistics Canada for 1996, the year of the most recent census.

The key independent variable relevant to the hypothesis was the number of low-fare carriers serving a given city-pair. As explained above, a low-fare carrier was defined as any airline (charter or scheduled) other than Air Canada or Canadian Airlines that served a particular city-pair. As will be seen, more detailed regression analyses were also performed to identify specifically which low-fare carriers were associated with reduced fare levels on the major carriers.

RESULTS

Table 1 shows the correlation matrix, and Table 2 shows the regression results. For each of the four regression models, all independent variables were entered simultaneously. An alternative model would be to use a two-step process of entering independent variables, in which the control variables (distance and population) were entered in the first step and the other variables were entered in the second step. When this two-step process was attempted, the same variables were significant as with simultaneous entry, and the Adjusted [R.sup.2]s were slightly lower. The simultaneous-entry model was used in Table 2 because it resulted in a clearer presentation.

Consider each of the four regression analyses shown in Table 2. Regression (1) shows that for fares on Air Canada or Canadian Airlines requiring a Saturday stayover, the significant independent variables are the distance between the two cities, the population of the larger city, the population of the smaller city, and the number of low-fare carriers serving that city-pair. This last variable is of particular relevance in this study. In other words, the presence of a low-fare carrier forces the major carriers to offer lower fares, and the effect increases as the number of low-fare carriers serving that route increase. Keep in mind that Regression (1) deals with fares requiring a Saturday stayover -- essentially the non-business traveler -- where the competitive impact of low-fare carriers ought to be especially strong.

However, note that the phrase "low-fare carrier" covers a variety of kinds of airlines, both scheduled and charter. Hence, Regression (2) is identical to Regression (1), except that the variable "number of low-fare carriers" has been broken down into five components. Four of the better known carriers were treated as separate independent variables (Air Transat, Canada 3000, Royal Airlines, and WestJet), while "Other Scheduled Carriers" constituted yet another independent variable. The last variable included AirSask, Athabasca Airways, Bearskin Airlines, First Air, Harbour Air, Helijet Airways, PemAir, and Tropical Airlines (no longer operating). For instance, if WestJet served a given city-pair, it was coded 1 for that city-pair; otherwise it was coded 0. Regression (2) shows that when "low-fare carriers" was broken down into smaller components, its effect was attributable to WestJet alone. WestJet was the only low-fare carrier whose service between a city-pair was associated with lower fares from Air Canada or Canadian Airlines. The other low-fare carriers appear to have had no significant effect.

Regressions (3) and (4) are similar to (1) and (2), except that the dependent variable is the unrestricted airfare between two city-pairs on Air Canada or Canadian Airlines. Here, one would expect the impact of low-fare carriers to be muted, since unrestricted airfares are very high and directed toward the business traveler, who is assumed to be insensitive to price. Yet here too, Regression (3) shows that the number of low-fare carriers was a significant independent variable. Surprisingly, its predictive power was greater upon unrestricted airfares than it was upon fares requiring a Saturday stayover. Regression (4) suggests this effect was attributable to Westlet alone among the low-fare carriers.

A caution should be noted in connection with Regressions (3) and (4): The number of city-pairs in the data set dropped from 163 to 107 for these latter two regression analyses. This is because the Expedia data bank lists the various fares available beginning with the cheapest, and if there are a particularly wide variety of inexpensive fares, the website doesn't have room to display the more expensive (unrestricted) airfares. Thus the 56 citypairs that were dropped from the analysis between the first two and the latter two regressions would be those city-pairs with the largest number of inexpensive fares. This introduces an element of bias into Regressions (3) and (4).

Strictly speaking, multiple regression can be used to demonstrate statistical association, but cannot prove causation. Nevertheless, it is assumed in this study that the causation is in the direction of low-fare carriers' presence affecting fares of major airlines. Reverse causation in this instance would mean that low-fare carriers were attracted to routes where the major airlines maintained fares at an unusually low level - not a tenable explanation.

DISCUSSION

In one regard, these results are not surprising, and confirm that just as Southwest Airlines exerts downward pressure on U.S. domestic airfares, a similar effect occurs in Canada where Westlet serves a city-pair. What is more surprising is that the phenomenon does not appear to carry over to other low-fare carriers such as Canada 3000, Royal Airlines, Air Transat, or the small scheduled carriers. It appears that the major carriers react to the entrance of WestJet into a market, but do not react to the entrance of other low-fare carriers.

Much like Southwest, WestJet has some unique properties that give it a competitive edge over the major carriers. WestJet's costs on short-haul flights are 40 percent lower than those of Air Canada. In part, this is because WestJet's s workforce is non-union and therefore more flexible. Also, WestJet uses only one type of aircraft, Boeing 737-200s, and signed a highly advantageous contract with a petroleum firm guaranteeing a price of US$18 a barrel until 2003 (albeit with a limit on quantity). [11] In March 2000 WestJet began service to two Ontario cities, with later expansion into Montreal and New Brunswick. Since WestJet's s success is predicated on the low costs of operating only short-haul flights, it will be interesting to see if the WestJet formula succeeds with a coast-to-coast network.

In a magazine interview, WestJet CEO Steven Smith was quoted as saying, "We're not competing head-to-head with Air Canada," and goes on to explain that the expense-account traveler, who's completely insensitive to price, is a lost market segment. [12] Yet Regression (4) shows that the major carriers reduced even unrestricted fares, normally sold to price-insensitive business travelers, when Westlet served a given route. It's hard to imagine a person on an expense account choosing a low-fare carrier for any reason. Nevertheless, these results suggest that WestJet does exert some competitive pressure on the major airlines, even for that market segment that can afford unrestricted tickets.

Although these business travelers are willing to fly on WestJet, they are lost to the other low-fare carriers, which typically offer less-than-daily service. This is consistent with the regression results, in which the major airlines reduce unrestricted fares when WestJet serves a city-pair, but not when other low-fare carriers serve a city-pair.

Regarding the non-business traveler, who is likely to book fares with Saturday-stayover requirements, the major carriers do not lower their fares in response to the presence of Canada 3000, Royal, Air Transat, or the small scheduled carriers. This may be because these low-fare carriers are seen as attracting incremental new business to the route -- people who would otherwise not fly at all -- rather than intruding upon the major carriers' existing clientele. Otherwise, it would be difficult to explain why their presence has no impact on major carriers' fares. Once again, WestJet is the exception, since the major carriers do react to its presence in a market.

There appear to be three market segments: First, there are some business travelers who are entirely insensitive to price and will not fly any low-fare carrier. Second, some travelers (business or leisure) will switch among WestJet and the major carriers depending on price, and surprisingly, this group includes some of those who buy unrestricted fares. Third, certain leisure travelers will fly Canada 3000, Royal, Air Transat, or the small scheduled carriers, or not fly at all, and are not potential customers of the major airlines. Only WestJet effectively exerts competitive pressure on the major carriers; the other low-fare carriers bring in incremental increases in passengers and are not treated as a threat by the major carriers.

The implication of these results is that a stronger version of WestJet, or more carriers with the characteristics of WestJet, are essential to maintain competitive pressure on the major airlines. The now-defunct Greyhound Airlines operated similarly to WestJet and fulfilled the same role; it was unfortunate for the consumer that it ceased operations, especially since Greyhound Airlines had a broader geographical reach than WestJet. Shortly after Greyhound's start-up, Canadian Business magazine [13] detailed its many difficulties, including regulatory battles due to foreign ownership, allegations of false advertising, and customer frustration because all long-haul passengers had to change planes in Winnipeg.

More new low-fare carriers would be desirable, but there is little government can do to force entrepreneurship in an industry that is so risky and that has such high start-up costs. An additional disincentive is that any small new airline would be vulnerable to predatory pricing by the major airlines, as has occurred in the U.S. Oum, Stanbury, and Tretheway [14] have documented other barriers to entry, such as restrictions on foreign ownership, control of customer reservations systems, shortages of gates at airports, and the impact of frequent flyer programs on customer loyalty.

Periodically there are fears that entrepreneurs may have given up on the airline industry entirely. In Canada, there has been a legacy of failed low-fare carriers during the 1980s and 1990s, such as Astoria, Nationair, City Express, and Wardair.

In December 1999, two separate proposals arose to start new airlines in Canada. Ken Rowe, owner of a large aircraft maintenance firm, announced plans to create a new discount carrier, "CanJet," which would have its hub at Hamilton, near Toronto. [15] When Air Canada heard of the CanJet venture, Air Canada hurriedly leased all available counter space at the Hamilton Airport, even though Rowe was in the midst of negotiations with the airport. [16] Although the federal government later required Air Canada to relinquish its control of counter space at Hamilton, Rowe eventually scuttled the proposed airline. At about the same time, Ken Deluce, whose family had owned a regional feeder airline, assembled a group of investors to buy up various regional feeder airlines from Air Canada and Canadian Airlines, combining them to form a coast-to-coast carrier named "Regional Airlines." [17] Deluce's proposal hinged on the federal government forcing Air Canada to sell its regional affiliates as part of its acquisition deal . This has not yet occurred and Deluce's proposal appears moribund. Both of these ventures were noteworthy because they were well capitalized and backed by people with experience in the industry. The likelihood of further new entrants into the Canadian airline industry is bleak indeed.

Following the extraordinary turmoil of late 1999, the future of the airline industry in Canada hinges on some yet-unanswered questions: How will Air Canada handle the power of being in a near-monopoly position? Will Canadian entrepreneurs continue to create new low-fare carriers on the WestJet/Southwest model when the barriers to success are so high? In the end, the Canadian government may have to choose between allowing cabotage or a return to the era of regulation.

Mr. Mentzer is associate professor in the College of Commerce, University of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 5A7. This article is revised from an earlier paper that was presented in the Tourism and Hospitality track of the Administrative Sciences Association of Canada 1999 Conference, Saint John, New Brunswick.

ENDNOTES

(1.) Transport Canada, Transportation in Canada 1997: Annual Report, Cat. #T1-10/1997E (Ottawa: Minister of Public Works and Government Services, 1997).

(2.) I. Lewis, "The Emerging Integration of the Canadian and U.S. Airline Industries," Transportation Journal, 35 (1996), No. 3, pp.49-55; T.H. Oum, W. T. Stanbury, and M.W. Tretheway, "Airline Deregulation in Canada and Its Economic Effects," Transportation Journal, 30 (1991), No. 4, pp. 4- 22; M.W. Pustay, "Competition and Concentration in Canadian-U.S. Transborder Aviation Markers," Transportation Journal, 38 (1999), No. 4, pp. 5-17.

(3.) Data are from the companies' 1998 annual reports and Moody's Transportation Manual 1997 (New York: Financial Communications Co., 1997).

(4.) W. N. Evans and I. N. Kessides, "Living by the 'Golden Rule': Multimarket Contact in the U.S. Airline Industry," Quarterly Journal of Economics, 109 (May 1994), pp. 341-366.

(5.) Competition in the U.S. Domestic Airline industry: The Need for a Policy to Prevent Unfair Practices (Washington: U.S. Department of Transportation, 1998).

(6.) K. Yakabuski, "Onex's 'Bold Step': One Big Airline," The Globe and Mail [Toronto], August 25, 1999, pp. A1, A7; P. Fitzpatrick, "Air Canada Closes in on Merger Victory: Canadian Surrenders," National Post [Toronto], December 6, 1999, pp. C1, C5; "A Chronology of the Five-Month Airline War," National Post [Toronto], December 6, 1999, p. C5.

(7.) P. Fitzpatrick, "Air Canada Raises Fares in Runup to Monopoly," National Post [Toronto], December 8, 1999, p. Al; I. Jack, "Foreign Airline Competition Possible," National Post [Toronto], February 18, 2000, p. A7.

(8.) Within the Microsoft Expedia website, the feature labeled "lowest published fares" was used to obtain fares. (http://expedia.msn.com - Accessed November 11, 1998). However, since the time of the study, that feature has been changed substantially. Data on which city-pairs were served by Air Transat and Royal Airlines were obtained from a website titled "Herman's Discount Airlines," which is a travel agency based in London, Ontario, that presents complete information on those discount carriers' routes (http://www.herman.on.ca/airline.htm - Accessed November 11, 1998). Data on which city-palm were served by WestJet were obtained from a booklet titled WestJet Fall 1998 Flight Schedule, distributed by that company.

(9.) In 1997, Transport Canada had described the industry as consisting of three tiers, with Air Canada and Canadian Airlines International comprising the first tier, and longhaul charter-type carriers as a second tier, with WestJet and various short-haul carriers making up the balance of the industry. From Transport Canada, Transportation in Canada 1997: Annual Report, p. 109.

(10.) "A number of formulations of the cities' population were attempted to ascertain which resulted in the highest Adjusted [R.sup.2] when regressed against the dependent variable (i.e., lowest major carrier fare available with a Saturday stayover): (a) Adding together the populations of the two cities in each city-pair, (b) multiplying the populations of the two cities in each city-pair, (c) using only the population of the larger city as an independent variable, (d) using only the population of the smaller city as an independent variable, and (e) using the population of the larger city and the population of the smaller city in the pair as two separate variables. Option "e" yielded the strongest predictive power and was therefore used.

(11.) P. Verburg, "Reach for the Bottom," Canadian Business, 73 (March 6, 2000), No. 4, pp. 43, 44,46, 48.

(12.) P. Verburg, "Reach for the Bottom," p. 46.

(13.) "B. Hutchinson, "See Spot Fly," Canadian Business, 69 (June 1996), No.7, pp. 91, 92, 94.

(14.) T. H. Oum, W. T. Stanbury, and M. W. Tretheway, "Airline Deregulation..."

(15.) "P. Kuitenbrouwer, "Eastern Canada Gets New Discount Airline," National Post [Toronto], December 22, 1999, p. Cl, C6.

(16.) D. Olive and P. Kuitenbrouer, "Airline Hopefuls Ask Ottawa to Require Sell-off of Regionals: Air Canada Unwilling," National Post [Toronto], December 7, 1999, p. C4.

(17.) "D. Olive and P. Kuitenbrouer, "Airline Hopefuls..."