Schaller Consulting
Customer-Focused Solutions

Issues in Fare Policy: The Case of the New York City Taxicab Industry

by Bruce Schaller

Presented at Transportation Research Board 1998 Annual Meetings
TRB Paper Number 981167

INTRODUCTION

Fare setting is perhaps the most delicate task faced by taxi regulators. Taxi fleet owners, often politically savvy and influential, exert strong pressures in favor of increasing the fare. On the other hand, the press and general public, citing service problems and large medallion license values, look askance on assertions that the taxi industry needs greater fare revenues. Regulators must decide whether to change taxi fares knowing that their action will be scrutinized by the public and will critically affect their relationship with the taxi industry and their ability to effectively regulate taxi owners and drivers.

An additional complication is the difficulty of the task. Taxi fare increases are generally based on analyses of industry costs and revenues. Yet there is no authoritative independent source for taxicab financial data. Much effort is required to verify data provided by the industry or collect data from outside sources such as insurance companies and car dealers. As a result, regulators often focus their efforts on the cost and revenue analysis and are able to spend little time on broader fare policy issues.

This paper addresses several of the broader policy issues: what goals can realistically be sought in setting fares? How do fare increases affect the quality and quantity of taxi service? How should medallion values figure into fare setting? Is tapping the "excess profits" evidenced by medallion values a viable alternative to increasing fares?

These issues are discussed in the context of New York City’s experience, synthesizing an extensive amount of data and research generated over the past decade on taxi costs and revenues, driver incomes and service quality. The discussion should be of interest to public officials, taxi regulators and the general public both because it relates to the nation’s largest taxi industry and because it addresses issues common to the regulation of taxis across the country.

ISSUES DISCUSSION

1. What are the goals of fare changes?

From the industry perspective, the objective of a fare hike is to increase revenues. From the public perspective, the goal is better service. Regulators tend to balance industry and public needs by seeking to improve service when increasing fares. The balancing occurred in New York City during each of the last three fare increases (1987, 1990 and 1996). Even if regulators do not explicitly address taxi service quality issues, quality is an inherent aspect of fare policy because good taxi service requires sufficient revenues to cover operating expenses and attract qualified drivers.

2. Do fare increases achieve the stated aims of increasing revenue and improving service?

This question is easiest to answer with respect to industry revenues. By definition, revenues will increase unless demand for taxi trips is elastic with respect to the fare, i.e., unless falloff in demand offsets the revenue gain from a higher price. Unfortunately, previous attempts to estimate fare elasticities have not produced statistically conclusive results. (1) (2) A recent econometric model for New York City, using data for 1990 to 1996 and taking into account the effect of the taxi fare and level of economic activity on taxi trip-making, indicates that demand is inelastic. The demand elasticity is estimated to be -.22, meaning that demand falls by 2.2 percent for every 10 percent increase in fares. (Author’s unpublished data. The adjusted R-squared for the model is .94 and the t-statistic on the fare coefficient is 7.75.) This result translates into a revenue increase of 16.0 percent from a fare increase of 20 percent (as in 1996), assuming no changes in economic activity.

Whether fare increases produce better service is more problematic but a very important question that goes to the heart of regulatory concerns. The public and press are understandably skeptical about assertions that increased revenues will be applied toward newer cars or more skillful drivers. (3) The counterclaim that fare increases simply fuel higher profits is a powerful argument against raising the fare. (4) (5) (6) If owners are currently able to utilize decaying cars or marginally able drivers, why would they not continue to do so and pocket the cash from a fare increase?

The best way to address this issue is to examine historical experience. The experience with vehicle and driver quality in New York City indicates that fare increases do not by themselves improve service.

Starting with a simple indicator, one sees that the taxi vehicle fleet became older, not younger, in the wake of a large fare increase in 1987. The number of cabs over four model years old increased from 2,102 to 3,305 in the two years following the 22 percent fare increase. (7)

Another key area affecting service quality is driver incomes. Driver pay is of utmost importance to retaining the experienced drivers who provide the best service. (8) Previous studies found that driver incomes rise after a fare increase but then inflation and lease fee increases erode the value of drivers’ take-home income. From 1986 to 1993, despite two fare increases, there was no net change in driver take-home inflation-adjusted income (9). It appears that as long as owners could fill their cabs with licensed taxi drivers, there was no impetus for higher driver incomes.

While fare increases have not improved vehicles or driver incomes, they have helped in a third area. When fares rise, demand for taxi trips falls somewhat, making it easier for passengers to hail a cab. An econometric model of taxicab availability, based on 1990 to 1996 data, estimates that the elasticity of service availability with respect to the fare is 0.28 (the adjusted R-square is .95, t-statistic is 7.28; author’s unpublished data). This translates into a 5.2 percent increase in availability after a 20 percent fare increase, assuming no change in total industry mileage or local economic conditions.

In sum, fare increases can achieve the basic objective of increasing industry revenue. Fare increases also improve service availability. But gains in other areas such as vehicle and driver quality are uncertain at best. Thus, if regulators seek to improve service through fare increases, they should accompany the fare increase with other regulatory steps to ensure success. (See (10) for discussion of policy options.) This has been the approach in New York City, most recently by regulations adopted in concert with the 1996 fare increase that require newer cars and that cap lease fees. (11)

3. How can fare policy help achieve adequate availability?

The fare’s impact on service availability is rarely considered in fare-setting. How can policymakers integrate availability considerations into fare policy decisions?

The first step is to set a goal for availability. The goal might be, for example, to ensure that nearly all passengers can obtain their desired trip when hailing a cab. Progress toward this goal can be measured through service refusal complaints. A significant number of service refusals means that some customers—disproportionately passengers taking trips that are less economically attractive to drivers (8)—are unable to obtain service.

The key to reducing service refusals is clearly to increase taxi availability. From the mid-1980s to 1992, the number of service refusals fell from 1,700-plus refusal complaints a year to 884 complaints in fiscal year 1992—at the same time that availability peaked at 500 million kilometers (310 million miles). Since 1992, trends have reversed; refusal complaints increased to 1,480 in fiscal year 1996 as availability tightened to 425 million kilometers (265 million miles).

As noted above, the taxi fare affects taxi availability. The econometric model of taxi availability indicates that two other factors are also important: economic conditions (which are outside the control of taxi regulators) and supply of service. [Detailed table on  availability, supply, economic activity and fare included in Web version only!]  The econometric results indicate that a 10 percent increase in supply, measured as total kilometers of operation, produces a 11.4 percent increase in available kilometers.

Suppose, then, that regulators sought to attain 1992 availability levels. What increases in supply and/or fares would achieve this goal?

Figure 1 shows different combinations of the taxi fare and supply that produce 480 million available kilometers (300 million miles), assuming 1996 economic conditions. As shown graphically, 480 million available kilometers would be produced by:

• Increasing taxi supply from 1.16 billion kilometers (722 million miles) to 1.29 billion kilometers (800 million miles), a 11 percent expansion. (Note that supply increased 2.3 percent in late 1996 and 1997 with the issuance of 267 new taxicab licenses. The net expansion above this would thus be about 8.7 percent.)
  
• Increasing the cost of an average trip from $6.73 to $10.34 in real dollars, or by 54 percent. This in addition to the 20 percent increase effective in March 1996.
  
• A 12 percent fare increase to $7.54 combined with a 7.9 percent supply increase to 1.25 billion kilometers (779 million miles)—or a 5.7 percent increase on top of the new licenses already issued.

Figure 1. Combinations of Taxi Supply and Fares that Produce 480 Million Available Kilometers (300 million miles)

Clearly, these prescriptions would be difficult to implement. Another large fare increase is probably out of the question, both in terms of public acceptability and because of the financial windfall it would produce for the taxi industry. A more attractive alternative is to expand supply but this is also difficult. Only one new batch of taxicab licenses has been issued since the 1930s—the 3.4 percent expansion (400 cabs) in 1996 and 1997. Due to environmental constraints and industry opposition this fairly modest expansion took a decade to win approval, and no further expansion is planned.

As a practical matter, then, econometric modeling showing how fares and supply affect service availability are problematic as a guide for policy-makers working under political, environmental and financial constraints. Results show, however, the importance of considering supply and availability during fare setting deliberations. These results provide policy-makers with a rational basis to evaluate how their actions affect the ease with which passengers can hail a taxicab.

4. How can the taxi industry need a fare increase with medallion prices well over $200,000?

Any discussion of taxi fares inevitably involves medallion prices. Medallion prices are the trading value of the limited number of taxicab licenses. Medallion values are by definition produced by excess profits, that is, profits above a fair return on capital invested in the business. It seems logical that rising operating costs could be covered by the excess profits instead of having to resort to a fare increase.

To evaluate this logic we can begin by calculating the level of excess profit that generates large medallion values, then assess whether this money could be used in place of a fare increase and what the impact of doing so would be on the industry and service quality.

First, how much profit do medallion values represent? Using the latest data available, taxi owners who lease their cabs to drivers on a long-term basis averaged $21,300 in profits in 1993. Medallion prices were about 10 times that amount, consistent with interest rates to buyers of about 10 percent. (7)

What does this profit represent in practical terms? Total revenues from fares and tips for cabs leased long-term averaged $93,500 in 1993. Thus, profits comprised 23 cents of every dollar passengers paid to drivers of these cabs. Profit levels are somewhat lower for other parts of the industry. For all industry segments taken together, profits were 17 percent of revenues in 1993. (7)

In theory, then, the 20 percent fare increase of March 1996 could largely have been funded by the profits that drive medallion values. Is this substitution a real option for taxi regulators? Could regulators draw on this money to avoid a fare increase?

Two obstacles are apparent. The first is an issue of fairness. Medallion values reached current levels over several decades. Owners who accrued a large share of the windfall gains left the scene long ago. Most current owners paid well over $100,000 for their licenses. (7) Should government act to wipe out that investment?

It can be argued that in paying such prices, medallion buyers knowingly took the risk that changes in government policy might reduce medallion prices. Against this argument lie two considerations. One is that those contemplating buying taxi licenses had no practical choice other than to accept this risk. From the standpoint of most current owners, the choice was between accepting the risk in becoming a taxi owner or not becoming an owner. The risk seemed all the more reasonable after the City government itself benefited from the medallion system by auctioning new medallion licenses to the highest bidders starting in 1996. Thus, for government to consciously curtail or eliminate medallion values raises serious fairness issues.

A second obstacle concerns the practicality of "funding" a fare increase with owner profits. How could regulators force owners to reallocate money from profits to operating costs or driver earnings? It is problematic to get owners to allocate new revenues to improving service quality. It would be far more difficult to force owners to reallocate revenue. The difficulty would be particularly acute among the large number of owners who are paying off loans used to purchase or refinance their medallion licenses. Regulators would be asking owners, in effect, to forego making loan payments in favor of buying new cars, lowering lease fees charged to drivers or other service-related use of the money. If successful, the effect would be convulsive to the industry. Lenders would foreclose on the licenses, which would be disposed of at auction to new owners, presumably at far lower prices. Who the new owners would be and the quality of service they would provide are unknowns.

In sum, it is easy to argue that the excess profits embodied by medallion values should be put toward better service. Doing so does not, however, seem to be a practical policy option.

5. Are trends in medallion prices a good indicator of the need for a fare increase?

One of the most common problems for taxi regulators is determining industry costs and revenues and the need for a fare increase. Perhaps medallion prices provide a ready index for the need for a fare increase.

The experience in New York City suggests that while medallion prices are a barometer of industry finances, prices are not a useful guide for fare setting.

Simple observation shows that medallion prices tend to reflect the industry’s financial fortunes. During recessions in the mid-1970s and early 1990s, values fell as demand for service dropped. Conversely, prices escalated during economic expansions in the 1980s and mid-1990s. (7) Thus, there is clearly a relationship between industry finances and medallion prices.

Medallion prices are not a very precise indicator, however, because they are influenced by factors other than industry profits. For example, a steep run-up in prices from the fall of 1993 to the spring of 1994 is difficult to explain purely from rising demand. Equally if not more important were changes in medallion loan financing terms. During this period, interest rates declined from 10 percent to 8.75 percent and the loan terms grew from a mix of 10 and 15 year loans to a standard of 15 years. The result, shown in Table 1, was that buyers could take out much larger loans—and thus afford the higher medallion prices—while making similar or lower monthly loan payments.

TABLE 1 Financing Terms for Corporate Medallion Loans, 1993-94

CONCLUSION

What guidance does the New York City experience over the past decade offer regulators and others considering taxi fare policy?

The foremost point is the importance of fare-setting to a variety of taxi regulatory issues. Fare-setting must go beyond simply determining whether a fare change is justified based on cost and revenue analyses. Fare setting should be undertaken in the context of a variety of larger policy issues such as driver and vehicle quality. Service quality will not improve simply because the fare is raised, but a fare increase can be harnessed to other regulatory measures to guide additional monies into new-vehicle purchases and higher driver incomes. For both substantive and political reasons, fare reviews are timely opportunities to initiate policies aimed at achieving these objectives.

Regulators should also evaluate impacts on service availability when considering fare increases. Availability is clearly an aspect of service quality, and in this case, a higher fare improves service.

These are important, constructive and useful aspects to fare policy-making. By contrast, debates over medallion prices contribute little to fare-setting decisions.

REFERENCES

1. Fravel, Frederic D. and Gorman Gilbert, Fare Elasticities for Exclusive-Ride Taxi Services. Prepared for the Urban Mass Transportation Administration, October 1978.

2. Hickling Corporation. Evaluation of Taxi and Limousine Service Demand and Economic Model for Taxi Rate Structure. Prepared for Regional Municipality of Ottawa Carleton, December 1990.

3. New York Times (editorial). The Taxi Industry Wins Again. December 10, 1989.

4. Roistacher, Elizabeth. The New York City Taxi Industry: What Price Medallions? City Almanac, Summer 1988.

5. Rogoff, Edward G. Don’t Raise the Taxi Fares. New York Times, October 30, 1995.

6. Rogoff, Edward G. Taken for a Ride by the Taxi Fare Hike. New York Times, March 28, 1987.

7. Schaller, Bruce. New York City Taxicab Fact Book, 3rd Edition. NYC Taxi and Limousine Commission, May 1994.

8. Schaller, Bruce and Gorman Gilbert. Improving the Proficiency of New York City Taxicab Drivers. Transportation Quarterly, Vol. 49, No. 4, Fall 1995.

9. Schaller, Bruce and Gorman Gilbert. Villain or Bogeyman? New York’s Taxi Medallion System. Transportation Quarterly, Vol. 50, No. 1, Winter 1996.

10. Schaller, Bruce and Gorman Gilbert. Fixing New York City Taxi Service. Transportation Quarterly, Vol. 50, No. 2, Spring 1996.

11. Perez-Pena, Richard. Taxi Board Votes Age Limits for Cabs, and a Rise in Fares. New York Times, January 26, 1996.

12. Taxi and Limousine Commission. Should the Taxi Fare Go Up? March 31, 1994.

Home | Consulting | Projects | Publications | Transit Web Sites | Taxi Research | Site Map | What's New