Peter Cappelli.

The changing system of airline industrial relations online

. (page 1 of 2)
Font size
QR-code for this ebook

HD28
•M4I4



*V.:-.




WORKING PAPER
ALFRED P. SLOAN SCHOOL OF MANAGEMENT



THE CHANGING SYSTEM
OF AIRLINE INDUSTRIAL RELATIONS

Peter Cappelli
with the assistance of
Timothy A. Harris

Industrial Relations Section

Sloan School of Management

Massachusetts Institute of Technology

SSM WP #1626-85 February 1985



MASSACHUSETTS

INSTITUTE OF TECHNOLOGY

50 MEMORIAL DRIVE

CAMBRIDGE, MASSACHUSETTS 02139



THE CHANGING SYSTEM
OF AIRLINE INDUSTRIAL RELATIONS

Peter Cappelli
with the assistance of
Timothy A. Harris

Industrial Relations Section

Sloan School of Management

Massachusetts Institute of Technology

SSM WP i*! 626-85 February 1985



Presented at the December meetings of the
Industrial Relations Research Association, 1984.



The authors are Assistant Professor and Master's student, respectively,
at the Institute of Labor and Industrial Relations, University of
Illinois at Urbana-Champaign. The authors are indebted to Jim Conway
and Jerry Glass of the Airline Industrial Relations Conference (AIRCon)
for providing access to the contract data analyzed below. This research
is part of a larger study on airline industrial relations conducted with
Robert B. McKersie and funded in part by the MIT Center for Transportation
Studies, Dan R. Roos, Director.



[^:a51t.




Airline Industrial Relations in Transition

Recent changes in collective bargaining in the airline industry stand in
marked contrast to the many years of stable industrial relations before
deregulation. The sharpness of these changes in industrial relations and
their correspondence with the restructuring of the industry caused by deregulation
provide an interesting example fron which to consider the economic forces
that govern industrial relations. In air transport, it is not the market
conpetition caused by deregulation as such that has changed industrial
relations but the fact that this conpetition was imposed on a system of
bargaining created for a different econonic environment. The pressures that
resulted have led to an elaborate pattern of lonion concessions and contract
changes aimed at restructuring airline labor costs. This pattern, especially
the differences across work groups, is best explained by examining the
characteristics of the different unions representing air transport workers
and the interests of their members.

In order to examine the pressures generated by deregulation, it is useful

to consider the situation that existed prior to the Airline Deregulation Act

1
of 1978. The first relevant aspect of the regulations imposed on the industry

by the Civil Aeronautics Board (CAB) was that carriers were not allowed to

add or withdraw flights fron their schedule without the Board's permission.

This restriction was designed to ensure an adequate, integrated air transport

network; where existing levels of service resulted in losses, the CAB was

authorized to pay subsidies to carriers in order maintain service.

The second relevant feature of the regulated environment was that the CAB

controlled fares, typically with formulas based on distance travelled, that

were uniform across carriers. Fare increases were granted by the CAB in

order to maintain reasonable rates of return, and wage increases were one of

the allowable costs that fare increases were granted to cover.



1. For an accoiont of the CAB controls prior to deregulation, see Taneja
(1976).



With respect to industrial relations, one cxDnsequence of this regulated
environment was that it gave the airline unions considerable power in bargaining.
Because fares could be raised to cover increased labor costs, the carriers
had less reason to resist then. Increased labor costs did not necessarily
put carriers at a price disadvantage, and CAB restrictions on schedules and
routes reduced the elasticity of demand so that higher fares led to relatively
small declines in traffic. The nature of air transportation and of collective
bargaining arranganents in the industry made it possible for the unions to
inflict considerable damage on the carriers through industrial action. As
Northup (1977) and Kahn (1980) point out, because it is impossible to inventory
passenger traffic, carriers lose all of the business they would otherwise
have had during a strike. Further, passengers stay away fran a carrier when
strikes are threatened and even after they appear to be settled, fearing sane
new dispute that will interrupt their travel. The craft nature of bargaining
under the Railway Tabor Act also meant that the different bargaining units
(pilots, flight attendants, mechanics, etc.), each had the ability to shut
the carrier down (Northrup 1971, Shils 1971).

In these situations, employers may feel that the cost of contract improvements
for these subgroups are small in absolute terms (affecting relatively few
workers) and are worth granting to avoid strikes (Kuhn 1971). Over time, the
different unions were able to whipsaw the carriers and drive total labor
costs up considerably. Hendricks, Feuille, and Szerszen (1980) found that
union wages and contract terms in the airline industry were higher than for
similar jobs in manufacturing, other things equal, in large part because the
environment created by the regulations increased union pcwer.

The carriers, hcwever, had two protections against the potential costs of
industrial action. The first was an industry strike fund, the Mutual Assistance
Plan, which became more generous over time and increased management's resistance



to strikes (Unterberger and Koziara 1977, 1980). The second protection came

fron the fact that the monopoly position that carriers often had in certain

routes under CAB regulations meant that after a strike, at least some of

their business would still be there.

The initial changes in the industry created by the Airline Deregulation

Act of 1978 seon to have increased union bargaining power even further. The

Mutual Aid Pact was legislated out of existance by the 1978 Act, and any new

strike fund new must meet a much more restrictive set of guidelines. In

addition, the end of CAB control over routes and schedules means that there

is now no guarantee that any of a carrier's business will be there after a

strike; coipetitors can cane into one's markets during a strike and take

1
those passengers away. At smaller carriers, however, the surfeit of pilots

and other skilled personnel during the recession made it possible for carriers

to threaten to break strikes by hiring replacements, possibly shifting sane

2
bargaining power back to managonent.

The increased vulnerability of the carriers to strikes and the threat

they may present to onployment has raised the stakes associated with industrial

action, and both sides are now extremely reluctant to engage in it. Indeed,

one of the main developments in industrial relations since deregulation has

been a very sharp drcp in strike activity. The most recent data suggests

that industrial action is at the lowest level in 16 years (NMB 1983), a

remarkable statistic given that the industry in general and labor



1. For example. United 's markets were apparently so severely damaged by its
58 day machinist strike in 1979 that it initiated half-fare coupons to try
and win sane of its business back; this move sparked the industry's first
major fare war which had disasterous consequences for all participants
( Forbes , 9/1/81).

2. Continental replaced its striking mechanics and pilots in 1983 and unilaterally
imposed lover pay rates as part of its bankruptcy reorganization plan. Kahn
(1980 p. 399) notes that Century Airlines took somewhat similar action in

1931. It took advantage of the surplus of pilots and the need to cut costs
and prices by forcing its pilots to resign and reapply for their jobs at half
pay. This action led to ALPA's first strike.



relations in particular are going through the most traumatic changes in their

history. In general, carriers have not won concessions through industrial

action; in many cases, they have agreed to substantial settlements rather

1
than confront unions determined to strike.

Of course, the most important change created by deregulation is that

carriers are new free to compete for markets on the basis of fares and

schedules. By itself, competition should not necessarily lead to pressures

for concessions; after all, the highest union wages and most stable industrial

relations have historically been in industries with ccmpetitive product

markets (steel, autos, manufacturing in general), seme of which were extremely

competitive (e.g., tires and meatpacking). Cctmrans (1909) famous argument

suggested that in order for unions to raise wages above the market level,

they must "take wages out of competition" by enforcing uniform contracts

across the entire product market so that no cdtpetitor will have a labor cost

advantage that can be turned into a competitive, price advantage. Where the

unions were able to do this, wages were protected no matter hcv/ competitive

the product markets were. In air transport, the unions have historically

covered virtually the entire product market. The major and national carriers,

all of which are at least partially unionized (even Delta has union pilots

and dispatchers), still fly over 90 percent of all revenue passenger miles;

the remainder goes to intrastate and "upstart" carriers, many of which are at

least partially unionized. The nonunion share of the air transport market is

roughly 5-7 percent, and these airlines often do not ccmpete in the



1. For example, 1VJA won a wage freeze from its pilots in 1983 but when
confronted by resistance from the machinists and flight attendants granted
them 30 percent increases (over three years) with no offsetting productivity
improvonents (BW 7/25/83). Some argue that Eastern caved-in to lAM pressure
last spring in order to avoid a strike despite a great need to secure concessions
(BW 10/10/83).



f

1

same markets with the tnank carriers. Clearly, this kind of market coverage
would be the envy of any industrial union in the country!

It might would seem reasonable, therefore, to conclude about deregulation
as did Hendricks et. al . (1980 p. 81) that "the industry and unionization
characterisitics that developed over 40 years of regulation have created a
bargaining environment that should not change substantially in the future. "
Then why has the pressure for union concessions been so relentless?

Despite their coverage of the product market, unions never enforced
uniform contracts across the product market and therefore never took wages
out of coTtpetition through collective bargaining. CAB restrictions on routes
and fares served that purpose, however, by preventing labor cost advantages
from being translated into lower fares and a corpetitive advantage. Because
of this CAB protection, there was no pressure forcing the evolution of
industry-wide bargaining of the sort that had occured in manufacturing. The
unions, therefore, directed their efforts toward other goals — meeting the
varying needs of members at the different carriers. They did this by giving
the locals almost complete autonony, especially in collective bargaining. As
a result, the bargaining structure in airlines has always been single craft-
single enployer. This type of bargaining structure was also encouraged by
the Railway Labor Act's requirement that representation be by craft,
leading to a plethora of unions in the industry. Shils (1971) points out
that significant industrial disputes in the industry generally involved only
one union, and disputes across carriers were virtually non-existant. This
bargaining structure remains despite the creation of a special coordinating
conmittee of air transport unions within the AFL-CIO.



1. These calculations are based on statistics on carrier market shares fron
CAB (1982). Freeman and Medoff's (1979) estimates suggest that 89 percent of
air transport production workers were covered by collective bargaining
agreements in 1969- '72. Coverage of the product market was virtually complete
because the organized carriers flew far more flights and typically did not
ccmpete with the nonunion carriers who were concentrated on intra-state
routes.



As soon as CAB regulations ended and fares became conpetitive, wages also

came under cxxrpetition. Because bargaining is carrier-specific, there is no

mechanism to prevent the different local unions fron undercutting each

other's labor costs. Financially-vulnerable carriers were able to secure

concessions and lower labor costs frcn their locals who were hoping to reduce

expected erployment losses; their competitors were then placed at a cost

disadvantage (one carrier estimated that 78 percent of their controllable

costs were labor related ( WSJ 9/29/83), so they also donand concessions.

1
Soon, the industry's wage structure cones apart.

As in other industries, unions in airlines agree to concessions as a way

to reduce potential unemployment; product demand has fallen because of rising

low-price coipetition (here, mainly fron other organized carriers), and the

labor danand curve which results suggests that fever workers will be employed

unless labor costs are reduced (Cappelli 1982, 1984). Sane carriers have

been pushed into banJcruptcy (Air Florida, Braniff, restructuring at Continental),

others have at least confronted th possibility (Western, Eastern), and many

carriers reduced capacity and laid-off workers during the recession. It is

not surprizing, therefore, to find financially-vulnerable carriers leading

the concession trend. Six carriers were rated as vulnerable during the post-

2
regulation period. All had secured concessions by the fourth quarter of

1981. Further, because of cash-flow and immediate financial problems, they

went after and secured wage concessions over workrule concessions by a margin



1. The great irony now is that during the early years of the industry, the
carriers had pushed for industry-wide bargaining that would have taken wages
out of ccmpetition but were rebuffed in these efforts by the unions. Brief
experiments in multicarrier bargaining with the lAM in the 1960 ' s were
abandoned (Kahn 1953, 1980).

2. The vulnerable carriers were Braniff, Continental, Eastern, Pan Am,
Republic, and Western. The ratings were based on a variety of financial
measxires including current assets to liabilities and cash reserves. The
financially-secure carriers were American, Delta, Frontier, Northwest, TVJA,
and United. More recently, IVJA could be seen as being in the vulnerable
group.



of 60 to 32. (Wage cuts generate immediate savings while workrule changes

cut average costs only down the road if operations expand or if workers can

be laid-off when operations contract. Unions are unlikely to agree to

workrule changes in the latter). In contrast, the six strongest carriers

financially during this period waited much longer before they secured concessions,

typically only after their ccrpetitors above had done so, and secured workrules

1
over wage concessions by a margin of 30 to 24.

There are many ways to reduce labor costs, and the contract concessions
secured by the carriers span a variety of areas in addition to wage cuts and
freezes. The most important concessions in the industry, especially for
flight cravs and attendants, concern schedules. (Indeed, for these groups,
contract restrictions on scheduling translate directly into pay through
credits and supplements for night duty, layover time, etc. See Baitsell
(1966) for a discussion of pay formulas.) About 45 percent of contract
concessions in 1981- '84 dealt with scheduling issues. In contrast to other
industries, there have been fewer efforts to broaden job classifications in
airlines presumably because of the resistance generated fron rivalries
between craft unions.

For many carriers, the issue has been whether contract concessions can
achieve the permanent restructuring of labor costs necessary to meet graving
corpetition frcm nonunion carriers who are currently hiring new enployees at
roughly half the pay of their more senior collegues at the trunk carriers.
The solution has been to introduce two-tier or 'B' wage scales which provide '
lower pay for new hires. Obviously, two-tier rates reduce average labor /



\



1. In many cases, however, labor cost differences are overshadowed by the
inportance of other operating characteristics (schedules, equipment, etc. ) in
determining a carrier's competitive advantage. For exartple, US Air has next
to the highest cost structure in the industry yet remains one of the most
profitable carriers; Braniff, in contrast, had next to the lowest cost
struct\are before going bankrupt. Further, the ability to secure concessions
from unions appears to be related to other considerations besides a carrier's
vulnerability. For example. United and American have been extremely successful
at securing concessions over time and are also among the industry's strongest
carriers financially; Western, Eastern, and others have been very vulnerable
financially but have over time been much less successful at securing concessions.
For a consideration of these relationships, see Cappelli ( f orthconing ) .



?



costs only as fast as the carrier can hire new workers — expanding the

workforce or at least generating turnover. For the unions, two-tier scales

represent a concessions that costs the current monbership nothing and which

creates incentives to hire new workers, helping to eliminate the problem

noted above of workers who have been stuck for years at the botton of the

seniority ladder. (The existance of two-tier scales raises potential problems

for union governance, however. ) American initiated the two-tier system as a

condition for future expansion in 1983. The rest of the industry hastened

to join the two-tier wave.

Two-Tier Wage Plans

AA DA EA FL NW OZ PI RC UA WA USAir
Pilots *1 *1 * * ? *i
Flight *i * * *i * * * * * *i
Attendants

Mechanics * * * * *

Agents 2***1 **2* *1
(Source: AIRCon data. 1 = new hire scales will not achieve parity with
scales for current employees; 2 = lower starting rates but same scales, ? =
negotiations in progress. AA=American; BA=Braniff ; COContinental ; DA=Delta;
EA=Eastem; FI>Frontier; NW=Northwest ; OZ=Ozark; PI=Piedmont; RC=Republic;
L!A=United; WA^estem. )

Perhaps the most interesting issue in airline industrial relations is the

distribution of contract changes by workgroup. Hew interested a workgroup is

in making concessions depends not only on the probability that concessions

will save jobs but also on the value of those jobs — hew do they compare to

^ alternatives elsewhere? In addition, the ability of local unions to grant

concessions may depend on their autonomy from the interests of the International

and on the extent of competition from other unions for their members. As

Ross (1948) argued, unions may feel compelled to take a harder line in

bargaining when they face competition from other unions. Together, these

arguments provide a good explanation of the pattern of contract concessions

outlined below.

Taken as a group, pilots have made more concessions than all other

workgroups combined. In almost every case, they have been the first group to

make concessions and have given up the most. The reason for this seons



clearly to be because pilots have the most to lose frcm layoffs. First,
alternative erployment with other carriers would result in a sharp pay cut.
Pilots who switch carriers lose their seniority and move to the bottcm of the
seniority pay scale at their new carrier. During this recession, as many as
5000 pilots were laidoff , suggesting that the ability to move to a new
carrier was renote. Second, there are almost no erployment prospects outside
of the airline industry that would make use of their skills. With respect to
union characteristics, many argue that pilots identify with management and
have more understanding of their problems than do other workgroups. In
addition, the trenendous autonony that the locals have in bargaining iirplies
that they are free fron pressure to maintain sane industry pattern. Further,
the fact that ALPA faces almost no competition frcm other unions seeking to
represent pilots gives makes it easier to take sometimes unpopular decisions
such as granting concessions.

The situation facing flight attendants is, perhaps surprizingly, quite
different frcm that of pilots. While there is no market outside of air
transport for these specific skills, flight attendants have less to lose from
layoffs than pilots first because their wages are considerably less and
second because seniority-based pay scales are less steep, making it easier to
move to a different carrier (CAB 1975). Perhaps most importantly, flight
attendants have historically had less attachment to their jobs than pilots;
if one is expecting to move to a different job, there is less interest in
making sacrifices to save the current one. The characteristics of flight
attendant unions also differ from the pilots. There are as many as 11 union
representing flight attendants, and the rivalry between them is intense.
Kahn (1980) noted, for example, that between 1976 and 1979, flight attendants
at six carriers changed their representation. As a result, the flight
attendant unions have taken much tougher lines in bargaining across the
carriers and have agreed to fever ( 18 percent of the total number ) , less
significant concessions than have the pilots.



/o



CO



c

3
O

u
o

o

:s

•o

c

0)

u
cu

•H

o
a

>^

CO

c
o

•H

m
m
cu
o

c
o
u



3

j«:
o

c

CO
(U

oo
to

^

o

(U

o

c

(U

•o

•H

o

c



4-1

a

D.
C/3






■r-r



J-J

4-J



I

X2






Si)



0)






CD

oc



(D



(1)

00



o



&
J









I ex



CN ,—1 Csl.— I CNrOCNl .


JiJ Sr>


3 H


Cu S


H ~~^


XI


a, XI



1—1



g



o
c



e-s



,8



CD
O







At the other end of the spectrum one finds the mechanics who have been
the workgroup far and away the least inclined to agree to concessions. Only
11 percent of the total nuiT±)er of concessions in the industry were granted by
mechanics, and they were typically far less significant changes than for
other groups. Frcm the employers' point of vie^, the labor cost differential
associated with mechanics is not great relative to the nonunion ccmpetition
because the mechanical work for the latter is typically done under contract
by the larger unionized carriers. Further, alternative ernployment is much
more available at other carriers and outside air transport (in manufacturing,
e.g. ) at wages corparable to those paid by the tnank carriers. Perhaps most
importantly, the structure of the lAM which represents the vast majority of
airline mechanics works to limit concessions. The International has the
ability to nulify local agreements and has used that power to prevent concessions
at individual carriers (H^ 4/9/84). The lAM has a strong incentive to avoid
concessions altogether in order to prevent then fron spreading to its negotiations
outside of air transport where similar settlement patterns are followed.

In many cases unions are able to secure inprovements in some aspects of
employment relations in return for granting concessions. These quid pro quo s
typically are secured in areas which do not raise current labor costs, often
expanding negotiations into new areas outside of the current contract.


1