Kennedale, Texas, appeals the district court's grant of summary judgment. We reverse and remand.

I. FACTUAL BACKGROUND AND PROCEDURAL HISTORY

This appeal raises a single question: Does the evidence offered by the city of Kennedale sufficiently support its ordinance regulating sexually oriented businesses?

In 1999, Kennedale annexed land that included multiple sexually oriented businesses, thereby subjecting those businesses to the city's ordinances. The ordinances prohibit the operation of sexually oriented businesses within 800 feet of churches, schools, residences, day care centers, parks, and other sexually oriented businesses, as well as within specified overlay districts. Additionally, the ordinances require sexually oriented businesses to obtain a license to operate. In justifying its ordinances, Kennedale relied on (1) studies from nine other cities, (2) an opinion survey of land use appraisers conducted by the city's attorney, and (3) citizen commentary from public meetings, all regarding the harmful secondary effects of sexually oriented businesses on surrounding land uses.

Following annexation, the ordinances allowed affected businesses three years to recoup their investments and relocate. Following criticism that the regulations failed to leave a sufficient number of alternative locations for already existing sexually oriented businesses, the city amended the ordinances to identify specific parcels of land upon which sexually oriented businesses may locate.

Reliable Consultants, Inc., d/b/a "Dreamers" (hereinafter "Reliable") is an off-site store, meaning that it sells video tapes, DVD's, magazines, and other print materials, but that none of the materials can be viewed or consumed on the premises, and the store offers no live entertainment, viewing booths, or theaters.1

After finding the ordinances were content neutral, the district court relied on Encore Videos, Inc. v. City of San Antonio, 330 F.3d 288 (5th Cir.2003), to find that the City's evidence of secondary effects failed to show that the ordinances were narrowly tailored to further a substantial government interest. The court declined to consider additional evidence Kennedale offered, and granted Reliable's motion for a permanent injunction. Kennedale appealed.

We review a district court's summary judgment ruling and other legal issues de novo. N.W. Enters. Inc. v. City of Houston, 352 F.3d 162, 172 (5th Cir.2003). We review a district court's factual findings for clear error. Kona Tech. Corp. v. S. Pac. Transp. Co., 225 F.3d 595, 601 (5th Cir. 2000). The Supreme Court's admonition that cities not justify ordinances by relying on "shoddy data or reasoning," City of Los Angeles v. Alameda Books, 535 U.S. 425, 438, 122 S. Ct. 1728, 152 L. Ed. 2d 670 (2002) (plurality opinion), requires factual findings, but turns on the legal interpretation of what the Supreme Court meant by "shoddy." Therefore, we review a district court's findings as to the existence of a city's evidence for clear error, but we review de novo whether that evidence falls within the Supreme Court's admonition.

"Zoning regulations restricting the location of adult entertainment businesses are considered time, place, and manner restrictions . . . if they do not ban [adult-entertainment] businesses throughout the whole of a jurisdiction and are `designed to combat the undesirable secondary effects of such businesses' rather than to restrict the content of their speech per se." Encore Videos, 330 F.3d at 291 (quoting City of Renton v. Playtime Theatres, Inc., 475 U.S. 41, 49, 106 S. Ct. 925, 89 L. Ed. 2d 29 (1986)) (citing Lakeland Lounge v. Jackson, 973 F.2d 1255, 1257-58 (5th Cir.1992)). Time, place, and manner restrictions on speech violate the First Amendment unless they are content-neutral, are designed to serve a substantial governmental interest, do not unreasonably limit alternative avenues of communication, and are narrowly tailored. See Encore Videos, 330 F.3d at 291-92.

Kennedale's ordinances meet the narrow tailoring standard if they "target[ ] and eliminate[ ] no more than the exact source of the evil [they] seek[ ] to remedy." Encore Videos, 330 F.3d at 293; Frisby v. Schultz, 487 U.S. 474, 485, 108 S. Ct. 2495, 101 L. Ed. 2d 420 (1988). Thus, an ordinance meant to deter property depreciation may only regulate businesses for which a connection to property depreciation can be demonstrated.

To show that an ordinance advances its goals, a city "may rely on any evidence that is `reasonably believed to be relevant.'" Alameda Books, 535 U.S. at 438, 122 S. Ct. 1728. However, "[t]his is not to say that a municipality can get away with shoddy data or reasoning. The municipality's evidence must fairly support the municipality's rationale for its ordinance." Id. at 438, 122 S. Ct. 1728.2

On-site businesses (i.e., adult theaters or strip clubs) pose a greater threat of secondary effects than off-site sexually oriented businesses (i.e., adult bookstores).3 Therefore, a city that enforces an ordinance meant to prevent harmful secondary effects associated with the operation of an off-site business must rely on evidence showing that off-site businesses, rather than the broader category of sexually oriented businesses that includes on-site businesses, cause harmful secondary effects. Encore Videos, 330 F.3d at 295 (requiring city to "provide at least some substantial evidence of secondary effects specific to adult businesses that sell books or videos solely for off-site entertainment" to meet narrow tailoring requirement).

In Encore Videos, we invalidated San Antonio's ordinance regulating sexually oriented businesses because the city failed to present adequate evidence showing a connection between off-site businesses and harmful secondary effects. San Antonio's evidence consisted of three studies conducted in other cities showing a connection between sexually oriented businesses, without isolating off-site businesses and secondary effects. Encore Videos, 330 F.3d at 294-95. Those studies did not provide any information exclusive to off-site businesses, so a substantial portion of the ordinance's burden on speech did not serve to advance its goals, and it failed the narrow tailoring prong. Id. at 295.

This case differs from Encore Videos because Kennedale, unlike San Antonio, offers evidence that purports to show a connection between purely off-site businesses, or "bookstores," and harmful secondary effects. To determine whether the ordinance at issue is narrowly tailored, we must determine whether Kennedale could reasonably believe that the evidence is relevant to show the requisite connection to harmful secondary effects. Alameda Books, 535 U.S. at 438, 122 S. Ct. 1728. In other words, we ask whether that evidence "fairly support[s] the [city's] rationale for its ordinance." Id. Applying our holding from Encore Videos, Kennedale cannot reasonably believe its evidence is relevant unless it sufficiently segregates data attributable to off-site establishments from the data attributable to on-site establishments. Encore Videos, 330 F.3d at 294-95.

Kennedale's evidence consisted of studies from nine cities, as well as an opinion survey of land use appraisers conducted by the city's attorney, and citizen commentary from public meetings. Seven of Kennedale's nine studies from other cities fail to differentiate between on-site and off-site businesses. The 1984 Indianapolis and 1986 Oklahoma City studies, however, included surveys of real estate appraisers that focused strictly on "adult bookstores." The overwhelming majority of survey respondents in both studies predicted that the presence of an adult bookstore would negatively affect real estate value in the surrounding area. The Indianapolis survey, conducted by the City of Indianapolis in conjunction with Indiana University School of Business, Division of Research, polled 20% of the national membership of the American Institute of Real Estate Appraisers.4 Eighty percent of the respondents predicted that an adult bookstore would negatively impact residential property values, and seventy-two percent believed commercial property value would also be negatively effected. The Oklahoma City study, which surveyed one hundred Oklahoma City real estate appraisers, produced similar results: Seventy-four percent predicted a negative impact on real estate value in the surrounding area.

Appellee Reliable argues that the term "bookstore," used in both surveys, is a term of art and does not sufficiently specify off-site premises. They argue instead that adult bookstores often include peep shows, arcades, and other forms of on-site entertainment, rendering them on-site establishments. However, the Supreme Court has previously used the term "bookstore" as distinguishable from "adult video arcades." Alameda Books, 535 U.S. at 442, 122 S. Ct. 1728 (discussing city's prohibition on "combination of adult bookstores and arcades"). This was a survey sent to and completed by real estate appraisers, and so what matters is how those appraisers would have understood the survey's reference to an adult bookstore.

Standing alone, it is reasonable to infer that the survey respondents interpreted "bookstore" as signifying an off-site establishment. Webster's Dictionary defines "bookstore" as "a place of business where books are the chief stock in trade." WEBSTER'S NEW INT'L DICTIONARY 253 (3d ed.1981). There is no reason to expect that simply adding the word "adult" to the term would completely transform the nature of the business activity described. Moreover, the Indianapolis survey also asked respondents to explain their prediction that an adult bookstore would negatively impact property value: 29% believed such an establishment would attract "undesirables" to the neighborhood, 14% felt it would create a bad image of the area, and 15% felt that it offended prevailing community attitudes. These reasons are equally applicable to an on-site or off-site establishment, and are distinguishable from the problems we have found to be unique to on-site businesses. See Encore Videos, 330 F.3d at 295 ("If consumers of pornography cannot view the materials at the sexually oriented establishment, they are less likely to linger in the area and engage in public alcohol consumption. . . ."). It is reasonable for Kennedale to believe that the appraisers responding to the survey understood the term "adult bookstore" to mean off-site businesses, such as that operated by the plaintiff-appellee.

Kennedale's ordinances purport to protect against harmful secondary effects. The Indianapolis and Oklahoma City studies support the belief that off-site sexually oriented businesses cause harmful secondary effects to the surrounding area in the form of decreased property value. So long as they are not relying on shoddy data or reasoning, we afford substantial deference to cities with regards to the ordinances they enact. See Alameda Books, 535 U.S. at 451, 122 S. Ct. 1728 (Kennedy, J., concurring) (noting that "a city must have latitude to experiment" and "courts should not be in the business of second-guessing fact-bound empirical assessments of city planners"). The Indianapolis survey, in particular, was drafted by experts, pretested, and administered to a large, national pool of respondents. It is not "shoddy." We therefore find that Kennedale has produced evidence that it could have reasonably believed was relevant, and thus could have properly relied upon. The ordinances are narrowly tailored to advance a substantial governmental interest.

The other evidence produced by Kennedale to justify its ordinance—an opinion survey of land use appraisers conducted by the city's attorney, and citizen commentary from public meetings—has also been hotly debated by the parties. Given our findings above, however, we need not reach that additional evidence. Similarly, our finding moots the question of whether the district court erred in excluding additional evidence of secondary effects.

By finding that Kennedale's ordinances were not narrowly tailored, the district court never reached the final element of the time, place, and manner analysis: whether the ordinances unreasonably limit alternative avenues of communication. We therefore remand this case to the district court to make those findings.

For the foregoing reasons, we REVERSE the district court's summary judgment and remand for findings as to whether the ordinances leave open sufficient alternative channels of communication.