Image Image Image Image Image Image Image Image

Center for Labor Research and Education


Scroll to top


Living Wage Policies at the San Francisco Airport: Impacts on Workers and Businesses

Industrial Relations, vol. 44, no. 1.

This paper evaluates the costs, benefits and related impacts of living wage policies implemented at the San Francisco Airport (SFO). Unlike other living wage ordinances, the policies at SFO cover a large proportion of the low-wage labor force in a distinct labor market. The authors find that about 73 percent of the ground-based non-managerial workers at SFO received substantial wage increases as a direct or indirect result of the policies; the proportion of these workers earning under $10 per hour fell from 55 percent to 5 percent, significantly reducing earnings inequality. Other benefits to workers included enhanced health benefits and an arrest of declines in quality of life indices. The costs of the policies to employers amounted to an average of 0.7 percent of fare revenue, or $1.42 per airline passenger. We observe a series of dynamic adjustments that reduced those costs, including dramatically reduced turnover, improved worker morale and greater work effort. We find some limited evidence of worker-worker substitution, but no evidence of employment decline.


Since 1994, living wage ordinances have been passed and implemented in over 100 local governmental entities in the United States. Most such ordinances establish wage and benefit standards in the $8.50 to $11 per hour range and some are indexed to local price indices. Typical ordinances also include incentives for employer payment of employee health benefits and provisions for paid  time off. Most living wage ordinances cover only employees on municipal service contracts; however, in some cities, the ordinances also cover employees working for employers who are themselves tenants on city-owned land or who are recipients of local business assistance tax subsidies; the number of employees actually covered by such clauses is thought to be very small. The early implementation of the ordinances typically involved the granting of numerous waivers and exemptions, which often reduced their impact. Consequently, the ordinances are thought to have small spillover impacts on the local low-wage labor market.

In 1999, San Francisco enacted a series of living wage policies, covering city service contractors, homecare workers and virtually all the low-wage workers at San Francisco International Airport (SFO). At the airport the most important of these policies–the Quality Standards Program (QSP)– affected about one-third of the 30,000 employees at SFO.1 The SFO policies, which represent one of the largest living wage experiments in the nation, are unusual also in that they included raised educational standards for new hires, training mandates that were intended to improve airport security and customer service, and also a large-scale labor peace/card check agreement. Given this great density of workers affected and the broader scope of these living wage policies, SFO provides a laboratory for observing the impact of grander labor market interventions.

Indeed, the SFO policies contain major components of what some commentators have labeled the “high-road” path of economic development.2 The high road program promises to raise low pay while also improving workers’ skills and productivity, and to reduce economic inequality while also enhancing economic performance. But can we get back on the high road?

This paper examines the impact of the living wage policies at SFO with these issues in mind. In the following section, we review recent research on living wage and minimum wage impacts. Next is a description of the scope and coverage of the living wage and related local labor market policies enacted at SFO. Then the research strategy for identifying the effects of the policy is discussed and the data sources collected to conduct a before and after comparison is described.

The main portion of the paper is devoted to identifying the benefits and costs of the policy and the dynamic adjustments that occurred. The benefits examined consist of the number of workers receiving pay increases, directly and indirectly, as well as additional health insurance benefits, paid time off and quality of life effects. The costs consist of the increased payroll costs to employers and the incidence of these costs. The dynamic adjustments that are examined concern first, changes in turnover, worker effort and productivity; and second, effects on aggregate employment and activity at SFO and specific effects on low-wage employment.

Related Research on the Impacts of Living Wages

Our research constitutes one of a new generation of living wage impact studies. This new research uses before and after comparisons of surveyed firms and workers in an individual living wage city. The papers by Brenner (2003) on Boston and by Fairris (2003) on Los Angeles in this symposium provide excellent examples of such work.3 Both Brenner and Fairris find substantial positive wage and turnover effects for covered workers.

These impact studies draw upon survey data we collected including firms that contract with the city as well as a control group of firms that do not. The advantage of this approach lies in this treatment and control design. The approach assumes that city service contractors and their workers who are subject to the ordinance are not a select group among the large population of firms in these industries. Fairris sheds some light on whether this assumption is valid for his analysis.

These papers constitute important advances in living wage impact research and represent the current state of the art. They are limited, however, in how much their methods can be applied to study the impact of policies such as those in effect at SFO. At SFO, spillover effects are likely to be larger and the treatment versus control methodology is consequently less applicable. At the same time, employment activity is derived from a single source activity. While city service contractors have the option of not providing services to the city, firms at SFO do not have the same choice not to play, so selection issues are less salient.

The recent literature on state minimum wage increases perhaps has greater relevance to our study, as state minimum wage laws typically cover most workers in one or more labor markets. The findings of Card and Krueger (1995) and Reich and Hall (2001) are especially pertinent because of their focus on California. Both studies drew upon Current Population Survey data.

Card and Krueger’s analysis of the 1988–1989 increase in the California minimum wage, presented in their 1995 volume, essentially compared California’s experience to a control group consisting of southern states that did not increase the minimum wage in this period. Card and Krueger found no measurable adverse employment impacts and some, although short-lived, real wage gains for low-wage workers. Using a similar difference-in difference methodology as Card and Krueger, Reich and Hall (2001) examined the impacts of the 1996 –98 California minimum  wage increases–from $4.25 to $5.75–by comparing employment trends in high wage and low wage industries in California. Reich and Hall found longer-lasting wage compression effects than did Card and Krueger and also could not detect any negative employment effects.

More recently, the California Budget Project examined the impacts of the 2001–2002 California increases–from $5.75 to $6.25 and then to $6.75– and found that employment in California grew faster than in the rest of the United States (CBP 2002). Indeed, from 1996 to 2002, California’s minimum wage increased nearly 60 percent, while at the same time California’s employment growth rate was higher than that of the rest of the nation–18.3 percent versus 12.6 percent.4

In sum, previous research on the impacts of living wages utilizes a treatment and control group design and finds generally that the benefits are substantial. However, this research is based on cities that have the very limited coverage of low-wage workers discussed above. The most pertinent recent minimum wage research on California examines a policy that has much broader coverage and also finds benign effects, but at mandated wages that are much lower than in typical living wage ordinances.

The Living Wage Policies at SFO: Scope and Coverage

Prior to the living wage policies, the employment and pay structure among SFO’s 30,000 workers was typical of most large U.S. airports. As Table 1 shows, passenger and cargo airlines accounted for approximately two-thirds of private sector employment, with the remainder concentrated among airline service companies (catering, security, skycaps and such aviation services as fueling and maintenance) and passenger service companies (retail and food concessions, airport parking lots and rental cars). Average pay growth in air transportation had lagged other sectors, including even retail, since airline deregulation began in 1978. Table 2 shows that the low-wage workforce at SFO was concentrated among the 11,000 ground-based, non-managerial workers, including: customer service and ramp workers, baggage handlers, screeners, cabin cleaners, and restaurant and retail workers.5 By 1999, over half of the ground-based non-managerial workers were paid less than $10 per hour.

Subcontracting accounts for a disproportionate share of the low-wage workforce.6 As Table 3 shows, employees of the airline service firms received lower wages and benefits. They received less training and had fewer opportunities for advancement than direct airline employees in the same occupational categories.

In an effort to raise wages and improve working conditions for workers, the San Francisco Airport Commission passed the Quality Standards Program (QSP) in January 2000. The QSP initially established a minimum pay standard of $9 per hour plus full health benefits, or $10.25 without health benefits, and it mandated 12 days per year of paid time off. It also established a high school completion hiring requirement and a training standard of 40 hours for new employees. It was fully implemented by October 2000. The QSP wage became $10 in January 2001, and is indexed for inflation.

The QSP covers ground-based workers who are employed in positions related to safety and security, generally those who work for the airlines or airline service firms. The following group of beneficiaries from the QSP are identified: United Airlines employees in the customer service, ramp and cabin cleaning divisions;7 all ground-based non-managerial employees of other airlines; and all non-managerial employees of airline services firms. Adding these groups together, there are approximately 11,000 workers in these jobs. In practice, this means that one-third of all airport workers were potentially affected by the living wage policies.

The QSP constituted only one of a related set of policies that substantially restructured the institutions regulating pay, benefits and labor relations policies at SFO between 1999 and 2001. The San Francisco Airport Commission and the San Francisco Board of Supervisors also passed a Minimum Compensation Ordinance (MCO), a Health Care Accountability Ordinance (HCAO) and a labor peace/card-check program. At the time of this study, the MCO and HCAO had only affected approximately 100 workers. While these policies cover most of the about 2700 ground-based non-managerial workers, including passenger service workers and employees of concession holders, not covered by the QSP, they go into effect only when leases are renegotiated.

The Labor Peace/Card Check policy did go into effect in February 2000; it had a substantial impact on labor relations at SFO.8 Approximately 2000 workers in the sample were organized into unions during the period of study. However, since the QSP set the general wage rate in collective bargaining agreements reached for workers covered by the program during the study time period, it will be regarded as the main policy that set wage and benefit standards at the airport.9

In summary, the scope and coverage of the living wage policies at SFO are much greater than in most other such ordinances. Moreover, as discussed in succeeding sections, the costs cannot be transferred to the city, as they can for service contractors. These parameters constitute an unusual set of conditions that distinguish the SFO case from most municipal ordinances, but which might contain valuable lessons for other area-based living wage ordinances, including those at large airports, and for citywide minimum wage ordinances.


  1. The 1999 living wage ordinances set comparable pay and benefits mandates for over 6000 workers in the homecare industry and an equal number of employees of the city’s service contractors. Together with the SFO group, these workers represented approximately half of all the workers in the U.S. who were covered by living wage ordinances in 1999. See Reich, Hall, and Hsu (1999) for more details. The focus in this paper is on SFO, as data on service contractor impacts was not yet available and Howes (2003) was conducting a parallel study of the impacts on the homecare sector.
  2. For “blueprints” of such proposed interventions, see Osterman et al. (2001).
  3. A study by Brenner and Luce (2003), which examines firm data for Boston, Hartford and New Haven, constitutes another example.
  4. For a more detailed discussion, see Reich and Laitinen (2003), as well as the survey article by Brown (1999).
  5. For additional details, see Reich, Hall and Jacobs 2003.
  6. In the 1980s, the airlines increasingly contracted out various ground-based services that used to be performed by direct airline employees. As a result, pay in these job categories declined. The authors review this evolution in Reich, Hall and Jacobs 2003.
  7. In this study, United Airlines was treated separately from other passenger airlines surveyed because of its dominant presence; United Airlines accounts for approximately half of all employment and airport activity at SFO.
  8. The Labor-Peace/Card Check policy requires employers operating at the airport to enter into card check agreements with unions that request such an agreement. Card check procedures call for immediate union recognition from the employer if a majority of employees sign cards. In exchange, the union agrees not to strike for recognition.
  9. Of course, this focus on the QSP was not always apparent to all the actors. They experienced all the policies, along with increased unionization and worker expectations that were altered by the union and living wage organizing campaigns as a totality that changed the economic and labor relations environment at the airport. Small sample sizes do not allow the authors to separate the impacts of these factors.