Opinion
LOW, P. J.
We hold that the San Francisco Transit Impact Development Fee Ordinance is valid.
In this consolidated appeal, plaintiff Russ Building Partnership 1 (hereafter Russ Building plaintiff), and plaintiffs Pacific Gateway Associates Joint Venture (Pacific), Crocker National Bank (CNB), and Crocker Properties, Inc. (hereafter collectively Crocker plaintiffs), appeal from a judgment upholding the validity of San Francisco’s Transit Impact Development Fee Ordinance (the Ordinance). They contend that the trial court erred (1) in [1503]*1503treating the Ordinance as a development fee and (2) in finding that the Ordinance could not be successfully challenged under articles XIII A and XIII B of the California Constitution, or the equal protection and due process clauses of the federal and state Constitutions.
In May 1981, the City and County of San Francisco (city) enacted Ordinance No. 224-81 “[i]n order to be able to provide public transit services for new development in the downtown area____” The Ordinance requires that the owners of buildings located in downtown San Francisco which contain newly developed office space, who have not yet received a building permit or a certificate of completion prior to the effective date of the Ordinance, pay a transit fee as a condition of issuance of a certificate of completion and occupancy. The fee is designed to provide revenue for the San Francisco Municipal Railway system (Muni) to offset the anticipated increased costs to accommodate the new riders during peak commute hours generated by the construction of new office space in the downtown area. The San Francisco Board of Supervisors fixed the fee to be paid by property owners at a maximum of $5 per square foot of new office space.2
Under the Ordinance, the transit fee is payable by each building’s owner in a lump sum at the end of construction. Alternatively, the owner may choose to amortize the fee over several years and make installment payments. The amount of the fee assessed is to cover increased transit costs which the city predicts will be generated over the 45-year life of each office building. There is no provision for adjusting the amount owed by each owner depending on the actual “life” of the building or actual transit operation costs. The board of supervisors may, however, adjust the $5-per-square-foot figure for future developers, depending on changing conditions.
In May 1981, Russ Building plaintiff filed a class action suit against the city to have the Ordinance declared invalid on its face and in its application. Another suit was filed against the city by Crocker plaintiffs, who additionally challenged the retroactive application of the Ordinance. The two cases were consolidated for trial of common issues.
The trial court entered judgment in favor of the city, finding that the Ordinance was not a tax and was a “debatably rational” development fee. It [1504]*1504also found that the retroactive application of the Ordinance as against the Crocker plaintiffs was legal. Appeals were taken separately from the trial court’s judgments and the cases have been consolidated for appeal before this court.
I
Both sets of plaintiffs argue that the transit impact fee is not a legitimate development fee because the $5-per-square-foot fee exceeds the reasonable cost of the increased services to be provided, and thus is a “special tax” which must be approved by two-thirds of the electorate pursuant to California Constitution, article XIII A, section 4 (hereafter referred to as section 4). Whether the transit fee is a development fee or a special tax is an issue of law. (See generally, Heckendorn v. City of San Marino (1986) 42 Cal.3d 481, 487 [229 Cal.Rptr. 324, 723 P.2d 64].)
Typically, a development fee is an exaction imposed as a precondition for the privilege of developing the land. Such fees are commonly imposed on developers by local governments in order to lessen the adverse impact of increased population generated by the development. (See Candid Enterprises, Inc. v. Grossmont Union High School Dist. (1985) 39 Cal.3d 878 [218 Cal.Rptr. 303, 705 P.2d 876] [school-impact fees]; Associated Home Builders etc., Inc. v. City of Walnut Creek (1971) 4 Cal.3d 633 [94 Cal.Rptr. 630, 484 P.2d 606, 43 A.L.R.3d 847] [dedication of land for park space or fee in lieu thereof].) This is one of the most common subjects of local police power regulations. (See Trent Meredith, Inc. v. City of Oxnard (1981) 114 Cal.App.3d 317, 325 [170 Cal.Rptr. 685].) “ ‘The position of the public entity is that this arrangement is only fair. The developer has created a new, and cumulatively overwhelming, burden on local government facilities, and therefore he should offset the additional responsibilities required of the public agency by the dedication of land, construction of improvements, or payment of fees, all needed to provide improvements and services required by the new development....’” (Ibid.)
The question which remains is how does the transit fee imposed here differ from a “special tax” within the meaning of section 4. The distinction between a tax and other exactments is admittedly blurred—taking on a different meaning in different contexts. Our Supreme Court has defined “special taxes” in section 4 “to mean taxes levied for a specific purpose rather than ... a levy placed in the general fund to be utilized for general governmental purposes.” (City and County of San Francisco v. Farrell (1982) 32 Cal.3d 47,57 [184 Cal.Rptr. 713, 648 P.2d 935].) (Heckendorn v. City of San Marino, supra, 42 Cal. 3d at p. 489.) In reaching this conclusion, the court held that the term must be strictly construed and ambiguities [1505]*1505resolved so as to limit the situations to which the two-thirds requirement applies. (City and County of San Francisco v. Farrell (1982) 32 Cal.3d 47, 52 [184 Cal.Rptr. 713, 648 P.2d 935].) The court reasoned that a restrictive interpretation was necessary because of the inherently undemocratic requirement that the tax must be approved by a supermajority of the electorate. (Ibid.)
While the Ordinance in this case does exact a fee for a specific purpose— to fund the cost of providing for increased ridership caused by the development—the crucial inquiry is whether the transit fee is a tax at all. The definition in Farrell makes the distinction between types of taxes only. To ascertain whether the transit fee is a tax, we first turn to the purpose behind section 4, using the strict interpretation approach announced in Farrell.
A
California Constitution, article XIII A was enacted to provide effective real property tax relief. “[Sjections 3 and 4 combine to place restrictions upon the imposition of such taxes.” (City and County of San Francisco v. Farrell, supra, 32 Cal.3d at p. 56.) One characteristic of a special tax is that it levies a fee to replace revenue for services which were affected by the reduction caused by article XIII A. (See Heckendorn v. City of San Marino, supra, 42 Cal.3d at p.489.) In contrast, the transit fee imposed by the Ordinance is not intended to replace revenues lost as a result of article XIII A. It is triggered by the voluntary decision of the developer to construct office buildings and is directly tied to the increase in ridership that this construction will possibly generate.
In Terminal Plaza Corp. v. City and County of San Francisco (1986) 177 Cal.App.3d 892 [223 Cal.Rptr. 379], the city enacted an ordinance which required residential hotel owners to make a one-for-one replacement of residential units which were lost through conversion or demolition as a condition for obtaining a building permit. In distinguishing this exactment from a special tax, the court concluded that “fees not exceeding the reasonable cost of providing the service or regulatory activity for which the fee is charged and which are not levied for general revenue purposes, have been considered outside the realm of ‘special taxes.’ [Citations.]” (Id., at p. 906.)
The transit fees required by the Ordinance were limited to the estimated costs involved to serve the increased ridership. None of the transit fees are earmarked for general revenue purposes. Further, unlike most taxes, the fees imposed by this Ordinance ■ re not compulsory but are exacted only if the developer voluntarily chor .xj , to create new office space. (See id., at p. 907; Trent Meredith, Inc. v. City of Oxnard, supra, 114 Cal.App.3d [1506]*1506at p. 328.) Developers have been required to pay for streets, sewers, parks and lights as a condition for the privilege of developing a particular parcel. There is little difference between these public improvements and the benefit to the public from the increased transit services paid for by the transit fee. In all instances, there is an increased burden on public improvements by virtue of the development. Moreover, such a construction will not interfere with giving voters effective property tax relief, the central purpose behind article XIII A of the California Constitution. (City and County of San Francisco v. Farrell, supra, 32 Cal.3d at p. 56.)
In Beaumont Investors v. Beaumont-Cherry Valley Water Dist. (1985) 165 Cal.App.3d 227 [211 Cal.Rptr. 567], the reviewing court had to decide whether a “facilities fee” imposed for connection to the local water district’s system was a special tax. The developer argued that the fee exceeded the reasonable cost of construction of the water system improvements required by the development and is a special tax. The reviewing court agreed. It relied on Government Code section 50076. That section was part of legislation enacted to implement article XIII A of the California Constitution. That section provides: “As used in this article, ‘special tax’ shall not include any fee which does not exceed the reasonable cost of providing the service or regulatory activity for which the fee is charged and which is not levied for general revenue purposes.” (Stats. 1979, ch. 903, § 1.) The court concluded that the people failed to sustain their burden of proving that the facilities fee did not exceed the reasonable value of the services for which it was collected. (Beaumont Investors, supra, at p. 238.)
Unlike Beaumont Investors, the city has undertaken numerous studies and has held public hearings to determine the reasonable cost of the increased transit services. To this end, the city hired consultants to project long-term needs and costs of transit services. Similarly, in J W Jones Companies v. City of San Diego (1984) 157 Cal.App.3d 745 [203 Cal.Rptr. 580], the city imposed a “facilities benefit assessment” on undeveloped property as a condition of obtaining a building permit which partially financed various public services built to serve the anticipated population generated by the development. These public improvements included sewer, street and transportation services, among others. The court concluded this was not a special tax but a special assessment levied on property which was benefited by the public facilities. (Id., at p. 758.)
Whether we term the transit fee a special assessment or a development fee, as applied in this context, the charge levied is directly related and limited to the cost of increased municipal transportation services engendered by the particular development, and it is not a “special tax” for purposes of section 4.
[1507]*1507B
Plaintiffs also contend that the transit fee is a “proceed of taxes” within the meaning of California Constitution, article XIII B and is subject to its limitations. That provision places limits on the growth of appropriations at both the state and local government levels, and requires excess appropriations to be returned by a revision of the tax rate or fee schedules within the next two years. (Cal. Const., art. XIII B, § 2; see County of Placer v. Corin (1980) 113 Cal.App.3d 443, 449 [170 Cal.Rptr. 232].) The “[p]roceeds of taxes” subject to the limitation is defined as “all tax revenues and the proceeds to an entity of government, from (i) regulatory licenses, user charges, and user fees to the extent that such proceeds exceed the costs reasonably borne by such entity in providing the regulation, product, or service____” (Cal. Const., art. XIII B, § 8, subd. (c).) Plaintiffs claim that the transit fee imposed by this Ordinance is a user charge or fee subject to the limitation. We disagree.
The transit fee is not a special tax within the meaning of section 4 and is not within the ambit of California Constitution, article XIII A, and therefore is not the type of revenue intended to be controlled by article XIII B of the California Constitution. Moreover, it is imposed as a condition of development of real property, which is not a regulatory license, user fee or charge. For these reasons, we conclude that the transit fee is not a “proceed of taxes” within the meaning of article XIII B. (See also City Council v. South (1983) 146 Cal.App.3d 320, 334-335 [194 Cal.Rptr. 110].)
II
Plaintiffs claim that the Ordinance discriminates against owners of office buildings constructed after 1979, denying them access to government benefits on the same footing as owners of pre-1979 buildings, and that it arbitrarily singles out commercial buildings while giving retail stores in the downtown area a free ride.
Under an equal protection analysis, the transit fee as an economic regulation is presumed to be constitutional. (Candid Enterprises, Inc. v. Grossmont Union High School Dist., supra, 39 Cal.3d at p. 890.) Since developers are not a “suspect class” and development is not a “fundamental interest” (ibid.; see also Trent Meredith, Inc. v. City of Oxnard, supra, 114 Cal.App.3d at p. 328), we determine only whether the “ ‘ “distinctions drawn by a challenged [Ordinance] bear some rational relationship to a [1508]*1508conceivable legitimate state purpose.” ’ [Citations.]” (Candid Enterprises, Inc. v. Grossmont Union High School Dist., supra, at p. 890.)
The stated purpose of the Ordinance is “to require developers of new [commercial office space] in the downtown area to pay a fee which is related directly to the incremental financial burden imposed upon the Municipal Railway.” (Ord. No. 224-81, § 38.3.) The impact fee is necessary because “[n]ew developments will bring increased need for public transit service in the downtown area during peak periods. The Municipal Railway will be burdened with the demands of transporting a larger number of passengers.” (Ord. No. 224-81, § 38.2.) The “Downtown Plan Environment Impact Report” indicated that the expansion of downtown office space is expected to exceed that of downtown retail space by 110 percent. The city concluded that “[f]uture increases in demand for public transit service” will therefore be “attributable directly to new development in the downtown area increasing the number of persons using the Municipal Railway during peak periods.” (Ord. No. 224-81, § 38.2.)
Consistent with the purpose of the Ordinance, the owners of new office space are required to pay for additional public facilities, the need for which is generated by their development. That the existing buildings will indirectly benefit from improved services does not result in such inequity as to offend equal protection principles. (See J. W. Jones Companies v. City of San Diego, supra, 157 Cal.App.3d at p. 757.)
Likewise, the argument that retail stores, which are also responsible for increased ridership, somehow got a windfall also fails. The Ordinance imposes the fee on the projected ridership directly and reasonably arising from the new office space. The city may rationally conclude that office workers increase the need for transit services during peak hours. The conclusion that it is office space, and not retail stores, that is primarily responsible for the need for improved transit services is properly left to the sound discretion of the local governing body. Since the trial court found this determination to be reasonably arrived at, this Ordinance must be upheld as being directly and additionally related to legitimate governmental goals. (Terminal Plaza Corp. v. City and County of San Francisco, supra, 177 Cal.App.3d at p. 910.) The trial court did not err in finding that the Ordinance does not violate equal protection standards.
B
Under a substantive due process analysis, “[a] law regulating or limiting the use of real property for the public welfare does not violate substantive due process as long as it is reasonably related to the accomplish[1509]*1509ment of a legitimate governmental interest. [Citations.]” (Id., at p. 908.) Plaintiffs maintain that it is unreasonable to force developers to underwrite public transit costs over a 45-year period.
There was no dispute that an office building has a useful life of 45 years and that over that lifespan it will generate additional passengers for Muni. Expert testimony demonstrated that long-term cost projections are employed in a variety of contexts, such as the purchase of capital equipment, financing pension funds and evaluating long-term bonds and stocks. Such long-term projections have been approved in deciding how much park and recreational space should be set aside to accommodate an increased population 40 years into the future. (See Associated Home Builders etc., Inc. v. City of Walnut Creek, supra, 4 Cal.3d at p. 639.) This is true even though inflation and other variables may affect the accuracy of such projections. Based on the data used by the city’s consultants, the trial court could properly conclude that the imposition of a lump sum fee representing increased transit costs over a 45-year period was not arbitrary or unreasonable and was not an unconstitutional taking of plaintiffs’ property.
Ill
Our Supreme Court has held that “[d]ouble taxation occurs only when ‘two taxes of the same character are imposed on the same property, for the same purpose, by the same taxing authority within the same jurisdiction during the same taxing period.’ ” (Associated Home Builders etc., Inc. v. City of Walnut Creek, supra, 4 Cal.3d at p. 642, quoting Rhyne, Municipal Law, p. 673.) In Associated Home Builders, it was determined that the requirement of land dedication for recreational use or the payment of “in-lieu fees” as a precondition of approval for development was sufficiently different from a property tax so that no unconstitutional double taxation resulted. This case is dispositive of the issue before us. The Ordinance is a development fee, not a tax. (See discussion, pt. I A, ante.) It is charged one time, at the completion of construction of new office space, and does not recur as does a property tax. Furthermore, the transit fee is designed specifically to fund Muni maintenance and development, whereas a property tax provides general revenue to cover a wide range of municipal services.
Plaintiffs attempt to distinguish the facts of the Associated Home Builders case on the ground that the fee imposed there was calculated differently than the fee in the instant case. Relying on Flynn v. San Francisco (1941) 18 Cal.2d 210 [115 P.2d 3], they maintain that the transit fee is more akin to a property tax because both the transit fee and property taxes are calculated based on the number of square feet in the development. This argument is without merit. In Flynn, the city imposed a “license fee” on certain types of [1510]*1510vehicles in addition to an ad valorem tax on these same vehicles. The license fee was unrelated to use of vehicles or any regulatory purpose. The court found that it was imposed only as an incident of ownership, as was the ad valorem tax, and constituted double taxation. (Id., at pp. 214-215.) Regardless of the method of calculation, the transit fee is not imposed by virtue of property ownership, but is a fee for the privilege of developing real property and to defer increased costs of transit services. As such, it is identical to the fee imposed in Associated Home Builders. As a matter of law, the imposition of the transit fee does not result in double taxation.
IV
The Crocker plaintiffs separately contend that the enactment of the Ordinance violates section 3.598 of the San Francisco City Charter. That section provides that “[rjates for each utility except the municipal railway shall be so fixed that the revenue therefrom shall be sufficient to pay, for at least the succeeding fiscal year, all expenses of every kind and nature incident to the operation and maintenance of said utility.” (Italics added.) If there is a deficit for “said utility,” section 3.598 of the San Francisco City Charter authorizes the imposition of a tax levy approved by a two-thirds vote of the board of supervisors. Under section 3.598, rates for Muni are proposed by the San Francisco Public Utilities Commission and are “approved, rejected or amended by the board of supervisors.”
The trial court found that the “charter section specifically excepts the Muni from the requirements of the section.” The trial court was correct in concluding that a Muni transit fee is not subject to the constraints of the above-cited language of section 3.598 of the San Francisco City Charter. Reviewing the plain language of the Ordinance, the board is not required to make up Muni deficits by levying a tax. The trial court correctly determined that the transit fee imposed does not violate the charter provision.
Having concluded that the city may constitutionally enact the transit fee pursuant to its police power, we offer no opinion as to the wisdom of the fee itself or upon the legality of any other exaction. We are mindful of the local government’s need to generate revenue to maintain the quality of life the residents of the city have come to expect. This has become increasingly difficult in the post-Proposition 13 era. The transit fee imposed falls within permissible bounds.
[1511]*1511y*
VI
Lastly, Russ Building plaintiff attacks the amount of the transit fee, contending that the methodology used and the assumptions relied upon by the city’s consultants were unsupported by the evidence. The board fixed the transit fee at $5 per square foot. This was based on a June 1981 estimate of increased transit costs calculated by Bruce Bernhard of the San Francisco Public Utilities Commission Bureau of Finance. This fee was less than Bernhard’s calculation of $6.57 per square foot. Prior to trial, the city retained Touche-Ross as a consultant. Touche-Ross calculated the increase in the transit costs to be $8.36 per square foot. This figure was based on a projected use over the next 45 years. The trial court found that the 45-year forecast was reasonable as it “rationally and directly related” to the estimated life of the buildings. The trial court also concluded that the assumptions and figures employed by the consultants in arriving at this amount were rationally based.
During the 10-week trial, each side presented exhibits and experts to testify about the methodology employed in calculating the transit fee. It is sufficient to state that the parties’ experts differed in their approach. Whether the approach taken by the city’s consultants was economically justifiable and financially and scientifically sound was a question of fact for the trial court, whose conclusion must be upheld if supported by substantial evidence. (See generally, United Business Com. v. City of San Diego (1979) 91 Cal.App.3d 156, 182 [154 Cal.Rptr. 263].)
Russ Building plaintiff challenges the use of the 45-year projection, claiming that it was not possible to estimate the increased transit costs that far into the future. Plaintiff relies on the testimony of its experts who stated, essentially, that for such a long period of time there were too many variables to consider to make any reliable estimate. Naturally, this conclusion was disputed by the city’s experts. While the degree of precision of the costs decreases proportionally with the length of the projection, we cannot state that such long term forecasts are inherently unreliable or are so arbitrary as [1512]*1512to be mere speculation. The trial court’s decision approving the 45-year term was supported by substantial evidence.
Next, Russ Building plaintiff argues that Touche-Ross used an improper discount rate in calculating the present transit fee. Touche-Ross applied a discount rate of 1.74 percent in arriving at the net present value of the lump sum fee levied against plaintiff. Plaintiff contends that the discount rate was too low and that it should be based on the rate for the “snapshot year” 1980-1981, the period from which the ridership data was taken. Touche-Ross used an economic forecast which estimated the discount rate over the 45-year projected term, rather than the rate for the “snapshot year.” This was the only long-term projection in the record, and there was testimony on both sides concerning its accuracy. The trial court found that there was a rational basis for using this method and we agree.
Russ Building plaintiff contends this figure is too low and that the discount rate for the snapshot year should have been used. We find no inconsistencies in the use of a snapshot year to predict the amount of increased services needed and the 45-year forecast to estimate the cost of those services over useful lives of the buildings. The increased usage of the transit system can be calculated to a reasonable certainty by selecting a model year, in this case 1980-1981, and extrapolating the need based on known quantities for ultimate office capacity, vehicle capacity, route structure and patronage volume.
In contrast to the future usage figure, the cost of those services is not reasonably ascertainable in any one year, but fluctuates with the economy. The cost for maintaining, repairing and purchasing new vehicles, and the costs of salaries and insurance, to name a few, are variable, subject to influences outside the control of city officials, e.g., gas prices, the inflation rate, etc. Under such circumstances, it would be unreasonable to base the discount rate on a “snapshot year” since it could be a year of unusually high or low inflation and, as we note, the inflation rate has not remained the same. In arriving at the cost of the increased transit services over the projected 45-year life of the building, it is more rational to use a forecast based on historical economic trends. We find substantial evidence to support the trial court’s finding.
C
Relying on Birkenfeld v. City of Berkeley (1976) 17 Cal. 3d 129 [130 Cal.Rptr. 465, 550 P.2d 1001], Russ Building plaintiff also argues that the [1513]*1513failure to include a rate adjustment mechanism is unconstitutional.3 Plaintiff proposes the city collect an annual installment, subject to yearly recalculations of the estimated cost. Plaintiff’s reliance on Birkenfeld is misplaced. In that case, a City of Berkeley rent control ordinance required a rollback of all controlled rents to those in effect on August 15, 1971; this would become the maximum rent subject to adjustments on a unit-by-unit procedure, which the court held was too cumbersome to be effective. {Id., at p. 136.) In this context, the court held that “[f]or such rent ceilings of indefinite duration an adjustment mechanism is constitutionally necessary to provide for changes in circumstances and also provide for the previously mentioned situations in which the base rent cannot reasonably be deemed to reflect general market conditions.” {Id., at p. 169.)
There is no similarity between the method for arriving at the maximum rent allowed in Birkenfeld and the projected costs for increased transit services in the instant matter. Unlike the Berkeley ordinance, the transit fee was not artificially limited to one particular year, but calculated based on the expected costs over the 45-year period, taking into consideration the fluctuating market conditions and inflation rate. A fair reading of Birkenfeld convinces us that an annual readjustment of the transit fee is not constitutionally mandated in this instance, and plaintiff’s contention necessarily fails.
D
Next, Russ Building plaintiff challenges the data used in the “snapshot year.”
First, plaintiff argues the city erred in assuming that no federal or state grants for capital improvements for expansion of the transit system would be issued. The 1984 amendment to the Ordinance provides that “capital subsidies ... shall be assumed only when and to the extent that receipt of such subsidies is reasonably probable.” (Ord. No. 224-84, § 38.5, subd. (b).) The trial court found that there exists “a debatable rational basis for concluding that State and Federal assistance for expansion vehicles and garage facilities will be unavailable . . . .”
In support of this finding, the trial court reasoned that since 1980 the city has had to use its own funds to finance all expansion vehicles. This evidence [1514]*1514went uncontradicted. Also, the agency responsible for approving transportation grants to the region, the metropolitan transit commission, treats expansion as its lowest priority, behind maintaining “existing and committed” service. While plaintiff demonstrated the city’s success with obtaining capital grants, there was insufficient evidence to show that the money received was used to expand upon existing facilities. On the basis of this record, we find sufficient evidence to support the trial court’s conclusion that the board could not reasonably anticipate government funds for expanding the transit system over the next 45 years.
Section 38.3 of the Ordinance requires the owners of the new buildings to pay the “incremental financial burden” imposed on the Muni system, and that the fee should “measure the total incremental burdens” generated by new office development. The 1984 amendment to the Ordinance requires that costs and revenues for offpeak periods not be included in calculating the transit fee. (Ord. No. 224-84, § 38.5, subds. (e), (f).) Plaintiff argues that it is illogical not to include revenues collected from offpeak service in offsetting the total increased costs of services.
Russ Building plaintiff argues that the city erred in ignoring the revenues generated during offpeak periods. The argument goes that since the new office workers will surely use the Muni during offpeak hours, the revenue collected therefrom should be used to offset the cost of the increased transit services.
The evidence reveals that Muni provides offpeak service at a huge deficit. Touche-Ross, the city’s consulting firm, used a marginal cost accounting approach and allocated the offpeak revenue to offpeak service, thereby reducing the deficit incurred during that period. There was no testimony that a surplus resulted from this accounting method. Thus, it does not appear that the increase in revenue during offpeak hours has any financial impact on the cost of funding an expansion of the system to accommodate peak service use. The trial court found that this approach was rationally based, and we agree.
Russ Building plaintiff also argues that the city deliberately understated the revenue to be generated by fast pass use during peak hours. In its statement of decision, the trial court found: Peak operating revenue is [1515]*1515composed of fast pass revenue and cash fare. In order to determine the average revenue per peak period trip, it is necessary to estimate the proportion of trips which are paid by fast pass versus the cash fare. Touche-Ross calculated that 54 percent of all peak period trips made by downtown workers will be paid for by fast pass, and 46 percent will be in cash. The trial court concluded that the estimate for fast pass use was a reasonable one.
Russ Building plaintiff contends that Touche-Ross mixed its data bases to arrive at an artificially lower revenue estimate from fast pass use. Touche-Ross combined the high number of fast passes sold at a discount price during 1980-1981 with the high estimate of monthly fast pass trips derived from a 1979 study in which the fast pass was not discounted. This, plaintiff claims, unreasonably inflated the fast pass/cash ratio and yielded underestimated revenue. We agree. In order to estimate the revenue likely to be generated by fast pass use, the calculation should be based on the cost per ride paid for at the time the ride is taken. If the Muni maintained a constant pricing policy, it would not matter which year would be used to derive the sales figures. But since the pricing of fast passes fluctuates, it is inconsistent to compare the last year’s cost of riding Muni with this year’s ridership. Plaintiff is correct in concluding that use of different periods artificially underestimates fare revenue derived from fast pass sales. The sales figures and the ridership use should come from the same target year.
Next, Russ Building plaintiff challenges the application of a “transfer rate” as part of the increased costs. Touche-Ross calculated a transfer rate to account for the times a rider is likely to transfer to another line in getting to his downtown destination during the peak period. The transfer rate was calculated to be 1.39, which meant that for every 100 trips to or from downtown, there would be 139 actual rides on the various Muni lines. The city’s policy, codified in Ordinance No. 224-81 section 38.3, provided that there be no increase in the level of crowding resulting from the new office space. The trial court did not question the “legislative wisdom” behind this policy and looked only to the reasonableness of Touche-Ross’s calculations. The court found the calculations to be reasonable. Plaintiff claims that the transfer rate arrived at is unjustified since the “crosstown” and “feeder routes” are not operating at or beyond capacity. We agree.
Touche-Ross calculated the transit fee so that the average “load factors”—a measure of the degree of crowding—does not rise due to construction of the new office buildings. The trial court found that even though the connecting routes were not operating at capacity, it was reasonable for the city to maintain the average load factors during peak periods because (1) it [1516]*1516could reasonably be expected that surplus space would disappear over the 45-year life of the project and (2) excess passenger capacity can be reasonably viewed as an asset to be preserved.
While it may be ideal transit policy to preserve the present load factors in the connecting routes, it is unfair to require Russ Building plaintiff to bear this burden alone. While the population generated by plaintiffs’ buildings will likely increase the ridership on the “feeder” and “crosstown” routes, there is no evidence of increased cost to Muni to operate these lines. This is the only relevant consideration, since the fee imposed by the city must not be more than needed to provide the improvements and services required by the development. (See Trent Meredith, Inc. v. City of Oxnard, supra, 114 Cal.App.3d at p. 325.) Anything beyond that would be unrelated to the development and constitute a tax. The trial court should have examined the city’s policy against the applicable constitutional standards. Had it done so, it would have concluded that only necessary expansion costs could be assessed against plaintiff.
However, we find these errors to be harmless. (See Warner Constr. Corp. v. City of Los Angeles (1970) 2 Cal. 3d 285, 301 [85 Cal.Rptr. 444, 466 P.2d 996].) The difference between the transit fee as originally estimated and the amount using the methodology discussed above would result in an estimated fee well above the $5 figure imposed on the Russ Building plaintiff. Using the Touche-Ross analysis relied on by the trial court, a recalculation of the fast pass revenue figure would result in a $0.60-per-square-foot reduction, from $8.36 to $7.76. (See fn. 2, antel) If the transfer rate were recalculated to exclude below capacity “policy lines,” the fee would be further reduced by 81 cents per square foot; from $7.76 to $6.95; still well above the $5-per-square-foot fee imposed by the city. Accordingly, it would serve no purpose to remand this matter for further findings.
VII
In summary, we hold that the transit fee was a lawful development fee which is not governed by California Constitution articles XIII A or XIII B, and it does not interfere with plaintiffs’ due process and equal protection rights. The fee is not a double tax and does not violate section 3.598 of the [1517]*1517city’s charter. Any error in calculating the fee was harmless, and the judgment in favor of the city and against Russ Building plaintiff is affirmed.
Each party to bear its own costs on appeal.
Haning, J., concurred.
See footnote, ante, page 1496.