NATIONAL ESTIMATES

Large attendance events can be subdivided into several event type categories. These categories include sport, entertainment, social, and cultural events. Information collection to develop estimates of the number of large PSEs nationally focused heavily on the individual event type categories. As mentioned earlier, the event type categories included in this report are:

1. Professional Football
2. Professional Baseball
3. Professional Basketball
4. Professional Hockey
5. College Football
6. College Basketball
7. College Hockey
8. Auto Racing
9. Horse Racing
10. Golf Tournaments
11. Marathons
12. Concerts
13. Parades
14. Fairs
15. Festivals
16. Protests/Political Events
17. Expositions and Shows

Several approaches were employed to collect event size and frequency data for these event type categories. The approaches included a venue and permitting authority-based approach and an association-based approach. The venue and permitting authority-based approach was designed to collect data from a sample of special events venues and permitting authorities, such as stadiums and police departments, at a micro or city-level. The results of this approach are provided in the previous four sections. The association-based approach was designed to collect data from trade associations representing the types of entities at issue within various special events categories at the national or macro level.

For the macro or national approach, the first step in the process was to collect the available data at the national level from the Census and from associations and other organizations. Data was collected by event type only for events of over 10,000 attendees. The information for these event types included number of events, total attendance, and in-event spending or revenue. Spending outside the venue, as well as indirect and induced economic effects, are estimated and applied separately later in this section. This process was effective for many types of events, particularly for sporting events which have national groups that collect such data, such as the NFL, NBA, and NCAA. However for other types of events, especially street-use events, data is often not available at the national level. Therefore, for these types of events, the local case studies were used to capture data and develop national estimates using scaling factors.

The remainder of this section includes 17 subsections, each of which discusses data collection for one of the event types. These subsections are followed by a discussion of the national economic and fiscal effects of PSEs and a summary of national PSE-induced congestion estimates.

Professional Football

The two major professional football leagues in the U.S. are the National Football League (NFL) and the Arena Football League (AFL). The NFL has the highest per-game attendance of any professional team sports league in the world.¹⁵ The NFL's 32 teams played 256 regular season games and 11 post season games in 2007.¹⁶ Average attendance at these NFL games was 68,301.¹⁷ The AFL hosts 144 games annually, with an average attendance of 13,105.¹⁸ Assuming all the AFL games had more than 10,000 attendees, the total attendance for both leagues was 20.1 million for 411 games in 2007.

There are several sources of data on spending and revenue for professional football. The 2002 Census reports revenues of $4.6 billion for 67 football clubs. However, the Census does not identify the clubs included in its estimates. Forbes magazine estimates NFL revenue in 2007 at $6.5 billion.¹⁹ Another source of data is the Fan Cost Index (FCI), an estimate of direct spectator spending on tickets, concessions and merchandise published in the 2007 Team Marketing Report (TMR).²⁰ The FCI includes two average-priced adult tickets and two average-priced children's tickets, two small draft beers, four small soft drinks, four regular hot dogs, two programs, two of the least expensive adult-sized adjustable caps, and parking for one car. The 2007 FCI for the NFL was $367.31. Accordingly, a per capita spending of $91.83 per attendee can be estimated by taking one quarter of the FCI. However, the FCI has been criticized by some for overestimating the average spectator spending on concessions and merchandise.²¹

Multiplying the FCI average spectator spending by the 18.2 million NFL regular and post-season attendees results in total spending of about $1.7 billion, far below the Census and Forbes revenue estimates. This gap is likely due to the large revenues from sources such as television, radio, advertizing, stadium naming rights, merchandise, and corporate suites.

The Forbes revenue estimate was selected as the best measure of the economic importance of the NFL, as it is more recent than the Census data and more inclusive than the FCI. As no estimates of AFL revenue were available, an estimate of AFL revenue per attendee was developed by taking the NFL's average revenue per attendee and applying it to the ratio of an average price of an AFL ticket to a NFL ticket. According to Team Marketing Report (TMR), the average NFL ticket is $67. The average price of an AFL ticket is $24.²² The ratio of the two average ticket prices is 0.35. Accordingly, average revenue per attendee for the AFL is estimated to be $128.23. Following this methodology, AFL revenue is estimated at $242 million. Adding the AFL revenue to the Forbes $6.5 billion NFL revenue estimate brings total professional football revenues to $6.7 billion.

Professional Baseball

The three major professional baseball leagues in the U.S. are the Major League Baseball (MLB) and the two Minor League Baseball AAA leagues (AAA). Among professional and college sports, MLB attracts the largest number of total attendees per season. This is due to MLB's long regular season schedule, currently 162 games per team with about 81 home games each, plus playoffs.²³ The MLB's 30 teams averaged 32,781 attendees per game in 2007.²⁴ It is assumed that virtually all MLB games had more than 10,000 attendees in 2007. The AAA hosts 366 games that would qualify as large PSE events, with an average attendance of about 11,500.²⁵ Between the two leagues there were 2,791 games with more than 10,000 spectators. These 2,791 games had a total attendance of 83.7 million.

There are several sources of data on spending and revenue for professional baseball. The 2002 Census reports revenues of $3.8 billion for 242 clubs. However, the Census does not identify all the clubs included in its estimates. Forbes magazine estimates MLB revenue in 2007 was $4.2 billion,²⁶ while the2007 FCI for the MLB was $176.55, or $44.14 per attendee. As previously noted, the FCI has been criticized by some for overestimating the average spectator spending on concessions and merchandise. It has been suggested that the average spending per attendee for MLB games may be 19 percent lower than the FCI.²⁷ Multiplying the FCI average spectator spending by the 79.5 million MLB season attendees results in total spending of $3.5 billion, which is lower than the Census and Forbes revenues estimates. This gap is likely due to revenues from sources such as television, radio, advertising, stadium naming rights, merchandise, and corporate suites.

The Forbes revenue estimate was selected as the best measure of the economic importance of the MLB, as it is more recent than the Census data and more inclusive than the FCI. As no estimates of AAA revenue were available, it is assumed that ticket and concession sales compose the majority of AAA revenue. Ticket and concession spending per attendee for AAA games is assumed to be $22.07, which is half the per-attendee spending for MLB games. Following this methodology, AAA revenue is estimated at $92.8 million. Adding the AAA estimates to the Forbes $5.48 billion MLB estimate brings total professional baseball revenue to $5.58 billion.

Professional Basketball

The two professional basketball leagues that host regular large PSEs are the National Basketball Association (NBA) and the Women's National Basketball Association (WNBA). The NBA has 30 teams and averaged 17,761 attendees per game for the 1,228 regular season games in 2007.²⁸ It was assumed that all NBA games exceeded 10,000 spectators in 2007. In 2007 the WNBA hosted 29 games with more than 10,000 attendees. The average attendance for those WNBA games was 12,654.²⁹ Between the NBA and WNBA leagues, there were 1,257 games with more than 10,000 spectators. These 1,257 games had a total attendance of about 22.2 million.

There are several sources of data on spending and revenue for professional basketball. However, the 2002 Census does not provide separate data for professional basketball. Instead, it is included in product line 7112119, Other Professional Sports Teams and Clubs, with 365 establishments and $4.7 billion in revenues. Forbes magazine estimates NBA revenue in 2007 at $3.5 billion.³⁰ The 2007 Fan Cost Index for the NBA was $281.90.³¹ A per capita spending of $70.48 per attendee can be estimated by taking one quarter of the FCI. However, multiplying the FCI average spectator spending by the 21.8 million attendees of the 2007 NBA season results in total spending of $1.5 billion, $2 billion less than the Forbes revenue estimates. This gap is likely due to the large revenues from sources such as television, radio, advertising, arena naming rights, merchandise and corporate suites.

The Forbes revenue estimates were selected as the best measure of the economic importance of the NBA, as they are more recent than the Census data and more inclusive than the FCI. In 2002, the total revenue of the WNBA was $85 million.³² Total attendance for all WNBA games was 1.7 million.³³ Accordingly, the average revenue per WNBA attendee was $49.93. Thus, in 2007 revenue from WNBA games with more than 10,000 attendees was assumed to be $18.3 million. The combined revenue of the NBA and WNBA for games with more than 10,000 attendees is approximately $3.6 billion.

Professional Hockey

The only professional hockey league that hosts regular large PSEs is the National Hockey League (NHL). Like the MBL and NBA, the NHL has 30 teams. The NHL averaged 16,957 attendees per game for 1,230 regular season games in the 2007 season.³⁴ All NHL games are assumed to have exceeded 10,000 spectators.

There are several sources of data on spending and revenue for professional hockey. However, like professional basketball, the 2002 Census does not provide data for professional ice hockey. Instead, it is included in product line 7112119, Other Professional Sports Teams and Clubs, with 365 establishments and $4.7 billion in revenues. Forbes magazine estimates NHL revenue in 2007 was $2.4 billion.³⁵ The 2007 FCI for the NHL was $282.95.³⁶ Per capita spending of $70.74 per attendee can be estimated by taking one quarter of the FCI. However, note that multiplying the FCI average spectator spending by the 20.9 million in attendance only results in total spending of approximately $1.4 billion, which is $1 billion less than the Forbes revenue estimates. This gap is likely due to alternative revenues from sources such as television, radio, advertising, arena naming rights, merchandise, and corporate suites.

The Forbes revenue estimates were selected as the best measure of the economic importance of the NHL, as there is no hockey-specific data in the 2002 Census and the Forbes estimate is more inclusive than the FCI.

College Football

Most large college sporting events in the U.S. are organized and regulated by the National Collegiate Athletic Association (NCAA), a voluntary association of about 1,200 institutions. The NCAA provides total season attendance figures as well as average attendance per-game estimates for the following college sports categories:³⁷

• Football
• Field Hockey
• Men's Soccer
• Women's Soccer
• Women's Volleyball
• Men's Basketball
• Women's Basketball
• Men's Ice Hockey
• Women's Ice Hockey
• Baseball
• Softball
• Men's Lacrosse
• Women's Lacrosse

The NCAA data does not provide attendance information on a per game basis, but rather provides information on total attendance, average attendance, and number of home games on a per-season basis. In order to develop PSE estimates for college sports, the number of games played by college teams with an average attendance per game of more than 10,000 attendees in each sports category was measured. While some of these teams' games had fewer than 10,000 spectators, those events were assumed to be offset by events of more than 10,000 spectators for teams with average attendance under 10,000. Of the college sport categories listed above, only four have teams with an average attendance of more than 10,000 attendees per game. All these teams were in Division I of their respective sports. The four categories are:

• Football
• Men's Basketball
• Women's Basketball
• Men's Ice Hockey

The NCAA has Division I, II, and III football programs. Division I football (which includes I-A and I-AA Subdivisions) has 152 teams with an average attendance of more than 10,000 people; Division II football has four teams that meet that criteria. Together these 156 teams played a total of 952 games and averaged 38,909 attendees per game in 2007. Division III football does not have any team that averages more than 10,000 attendees per game.³⁸

An NCAA report on the 11 major athletic conferences and a group of independent universities estimates the revenue for Division I-A football teams was $11.4 million per school in 2002.³⁹ Applying this revenue figure to all Division I and II football teams with an average attendance of more than 10,000 spectators results in a revenue estimate of about $1.8 billion for the 156 teams.⁴⁰

College Basketball

College basketball in the U.S. is organized and regulated by the NCAA. The NCAA includes Division I, II, and III basketball programs. Total season attendance in 2007 for Division I basketball was about 18 million for men's basketball and 4 million for women's basketball.

The most recent college basketball revenue estimates are from an NCAA report on the 11 major athletic conferences and a group of independent universities. The report estimated the average Division I men's basketball revenue per team at $3.9 million in 2002.⁴¹ The same report estimated women's Division I basketball generated $336,000 in revenue per team. The universities included in the NCAA report are assumed to correspond closely with the 100 men's teams and 50 women's teams for which the NCAA provides online attendance statistics. Accordingly, season revenue for the top 100 men's Division I basketball teams is estimated at $390 million and season revenue for the top 50 women's Division I basketball teams is estimated at $16.8 million.⁴² The NCAA does not provide data for the remaining Division I men's and women's teams. Teams for which the NCAA does not provide data were excluded from this study.

Average revenue per attendee was estimated using the ratio of total season attendance for teams with an average attendance of more than 10,000 spectators per game to the total attendance for all teams for which the NCAA provides online statistics. Division I basketball has 47 men's teams and 4 women's teams out of 325 that have an average attendance of more than 10,000 attendees.⁴³ The 51 men's and women's teams with average attendance above 10,000 played a total of 877 home games in 2007. These games attracted 62 percent of all attendance to the top 100 men's team games and 19 percent of all attendance to the top 50 women's team games. Therefore, men's and women's team games with average attendance of more than 10,000 had season revenues of $243.6 million and $3.1 million, respectively. The resulting revenue per attendee estimates are $21.68 for these men's teams and $4.17 for these women's teams. The weighted average revenue per attendee is $20.59.

College Ice Hockey

There are only three Division I men's ice hockey teams with an average attendance of more than 10,000 people per game. These three teams play a combined total of 68 home games each season and attract an average of 10,898 attendees per game. None of the Division I women's ice hockey teams has an average attendance of more than 10,000 spectators.

The NCAA does not publish college ice hockey revenue estimates. In order to develop revenue estimates for college ice hockey, revenue per attendee was assumed to be the same as men's college basketball. Assuming men's college ice hockey and college basketball have the same average revenue per spectator of $20.59, the total revenue of college ice hockey games with more than 10,000 spectators is about $15 million.

Horse Racing

The 2002 Economic Census includes data for 121 horse racetrack operations, with $3.8 billion in revenue. However, a national database of horse racetrack attendance and revenue is not available. California is the largest state in terms of racing revenue, and therefore, data on attendance and revenue from the California Horse Racing Board, along with Census data on the share of racing in California, was used to develop national estimates.

According to the California Horse Racing Board, in 2006 there were 944 horse racing events in California, including 279 with more than 10,000 in average attendance.⁴⁴ The average attendance at horse races with more than 10,000 spectators was 17,345. The total revenue from those races was $485.5 million, or $100.33 per spectator. While some of the races at racetracks with average attendance of more than 10,000 spectators have attendance of less than 10,000 spectators, those events are assumed to be offset by events of more than 10,000 spectators at racetracks with average attendance under 10,000.

Data from the 2002 Census indicates that California accounted for 9.9 percent of racetrack spectator revenues nationally in 2002.⁴⁵ The Census data includes auto, horse, and dog racetracks. Assuming the ratio is the same for horse racetracks, which are the most common type, there are an estimated 2,819 large horse racing events in the U.S. The California data for average attendance and revenue per spectator were used to estimate attendance and revenues at the national level. The total attendance at large horse races nationally is estimated to be 48.9 million, and the associated revenue is estimated to be $4.9 billion.

Golf Tournaments

Top-level professional golf consists of a year-round schedule of weekly tournaments. Most of the tournaments are organized into series, or tours. There are separate tours for men and women. Each tour is based in a specific geographical region in the U.S., though some of them also feature events in other parts of the world.

The major tours in the U.S. are operated by the Professional Golf Association (PGA) and Ladies Professional Golf Association (LPGA). The PGA organizes three major tours. The major tours of the PGA are the PGA Tour (54 events), the Nationwide tour (30 events), and the Champions tour (33 events). Fourteen of these events are held outside the U.S. and were excluded, reducing the number of applicable events to 103. The LPGA has 35 events. Among these four tours there are a total of 138 major golf tournaments. Golf tournaments are typically one week long, but usually have the highest attendance over the four day period between Thursday and Sunday, when the main event is played. These four days are assumed to attract at least 10,000 spectators at each of the 138 tournaments. This results in an estimate of 552 event days with more than 10,000 spectators nationally. While it is likely that some of these event days account for less than 10,000 spectators, non-tour event days at events such as the U.S. Open and the Seniors PGA Championship tour are assumed to offset those events.

Average or total attendance data for golf were not available. As a result, estimates were based on the average daily attendance of 22,125 for golf tournaments in the Portland case study. This attendance estimate appears reasonable against other data for some of golf's premier events. For example, the Saturday attendance for Tiger Woods' signature tournament in 2007 had 38,000 spectators. Additionally, the U.S. Golf Association usually prints 30,000 to 35,000 tickets for the U.S. Open, depending on the size of the venue.⁴⁶

Revenue data for the tours were also not available. However, the PGA and LPGA do publish the total purse information for each tournament. As these purses would represent the largest cost component of an event, and were the only monetary data available, they were used as a surrogate for economic magnitude. The purses of the four tours for events in the U.S. totaled $383 million, or approximately $31.37 per attendee.

Auto Racing

Auto racing is subdivided into many categories based on auto types. These auto type categories include:

• Single seater
• Touring car
• Stock car
• Sports car
• Drag
• Rallying
• Production car

Within these auto racing categories are multiple leagues that represent various regions, league levels, and auto specifications. For example, drag racing has two major leagues in the U.S., the National Hot Rod Association (NHRA) and the International Hot Rod Association (IHRA).

The 2002 Census includes data for 478 auto racetracks that have a combined $2.0 billion in annual revenue. However, neither the Census nor any other data source collects auto racing attendance statistics for all the auto racing categories. As a result, national auto racing attendance estimates were developed by contacting the major auto racing leagues in the U.S. The major auto racing leagues that were contacted include:

• National Association for Stock Car Auto Racing (NASCAR)
• IndyCar
• Grand American Road Racing Association (Grand Am)
• National Hot Rod Association (NHRA)
• International Hot Rod Association (IHRA)
• American Le Mans (ALMS)

According to the NHRA, these leagues are likely to be the only auto racing leagues that regularly host events with more than 10,000 spectators.⁴⁷ Data on the number of event days in all the major auto leagues was collected, as were most attendance estimates. However, attendance data for two major racing series were unavailable. These two series were the NHRA's POWERade and the Grand Am's KONI Challenge. Average race-day attendance at these two series was estimated using the average attendance of the known race attendances. Incorporating the POWERade and KONI Challenge estimate, there are an estimated 284 auto racing event days with more than 10,000 attendees, and the average attendance for these event days is 58,160 people. Detailed attendance estimates for the major leagues and series are provided in Exhibit 7-1.

Average in-event spending for auto races was estimated at $149, based on the Grand Prix data in the San Jose report described in the Study Methodology section.⁴⁸ This updated estimate was applied to all auto racing events to develop a revenue estimate of $2.4 billion.

Marathons and Walkathons

There is limited public information available regarding attendance and revenues for walkathon events. Marathon attendance information is more readily available. Each year, MarathonGuide.com publishes a report detailing national marathon attendance data. This report is claimed to be the most comprehensive and accurate marathon data available.⁴⁹ However, the report only records the number of runners that finished the race at each marathon and does not include estimates of spectators.

According to MarathonGuide.com, there were eight marathons with more than 10,000 participants in 2007. Assuming there are four spectators for every runner at a marathon, there would be 33 marathons with more than 10,000 participants and spectators. This assumption is based on the spectator-to-participant ratio of four to one at the 2006 Marine Corps Marathon in Washington, D.C.⁵⁰ Accordingly, the average number of participants and spectators at the 33 large marathons was 43,522. The average in-event spending per attendee at marathons and walkathons is estimated to be $49, based on data for the Rock n' Roll marathon from the San Jose study.⁵¹ Thus, marathons with more than 10,000 attendees generated $70.4 million in spending.

Attendance and revenue data for walkathons is very fragmentary and limited. Most walkathon organizers fundraise for specific causes such as health, poverty, and education. These types of walkathons are also called charity walks. A list of some major walkathons in the U.S. includes:

1. MS Challenge Walk - National MS Society
2. Stop the Silence - Prevention and treatment of child sexual abuse
3. Avon Walk for Breast Cancer - Breast cancer
4. Relay for Life - American Cancer Society
5. Race for Hope - Brain Tumor Society and Accelerate Brain Cancer Cure
6. Walk to Cure Diabetes - Juvenile Diabetes Research Foundation
7. Walk to Empower - Y-ME Nation Breast Cancer Organization
8. Susan G. Komen National Race for the Cure - Breast cancer
9. Ivy Vine Charities Walk for Diabetes and Scholarship
10. Take Steps Walk for Crohn's and Colitis
11. Walk for the Kids
12. American Kidney Fund Steps that Count
13. AIDS Walk
14. Down Syndrome Association Buddy Walk
15. National Kidney Foundation Kidney Walk
16. Race for Humanity - Habitat for Humanity
17. Step Out to Fight Diabetes - American Diabetes Association
18. Help the Homeless
19. Jingle Bell Run/Walk - Arthritis Foundation
20. March for Babies - March of Dimes Foundation

Race for the Cure and Relay for Life estimate all of their events are attended by more than 10,000 attendees.⁵² However, only Relay for Life provided specific attendance estimates for its events. Race for the Cure hosts 122 events with more than 10,000 attendees annually. The average attendance at these walkathons is 30,000 people. Based on the assumption that each of the 20 identified major walkathons hosts the same number of events as the Race for the Cure, it is estimated there are approximately 2,440 large walkathons in the U.S. annually. Average attendance at all these walkathons is assumed to be 30,000 per event, based on the data for Race for the Cure.

The spending per attendee estimate for walkathons was also $49, based on the San Jose marathon. When applied to walkathons with more than 10,000 attendees, this estimate indicates that walkathon events generate $3.58 billion in revenue annually. Combined with the $70.4 million estimate for marathons, this category generates $3.6 billion annually.

Concerts

Comprehensive national data on concerts across the U.S. is available from Pollstar, a weekly magazine and prominent trade publication for the concert industry. Pollstar estimates there were approximately 1,867 concerts in the U.S. with more than 10,000 attendees in 2007, and the average attendance at those concerts was 18,686.⁵³ Given these figures, total attendance at these large concerts is estimated to total 34.9 billion. The Pollstar data only includes ticketed concerts that charge an admission fee; free admission concerts are omitted. Unfortunately, there are no national estimates of free admission concerts in the U.S., and the case study cities did not provide enough information to generate national estimates. As a result, free admission concerts are not included in this report.

No national information on revenues from concerts was available. However, ticket revenue is the largest component of total concert revenue, and the average concert ticket price is $50, according to Pollstar. Based on this average ticket price, large concerts are estimated to generate approximately $1.7 billion annually.

A report by the National Endowment for the Arts entitled "The Performing Arts in the GDP, 2002" found that "consumers spent $12.1 billion ($42 per person) on admissions to performing arts events. This amount was $2.5 billion more than spending on tickets to movie theaters, but $1.5 billion less than outlays on admissions to sporting events."⁵⁴ The National Endowment for the Arts' estimate includes all types of performing arts events regardless of attendance. Large concerts represent only a small portion of this spending. Hence, the Pollstar data is considered to be more relevant to the purposes of this study.

Parades

Searches at the national level revealed very little comprehensive data on parade attendance and revenue. This was also true for other park and street type events including fairs, festivals, protests, political events, expositions, and shows. This is due to the wide dispersion in occurrence, large number of event organizers, and lack of a national association or trade group that collects information regarding these types of events.

To circumvent these challenges, national estimates for these types of events were developed using data from the case studies of four cities conducted as part of this research. For the parades case study data extrapolation, the results for each of the case study cities were scaled up based on the number of cities in their respective size class. The underlying methodology used to develop the scaled-up estimates from the case study cities is discussed in the Study Methodology section. Using this methodology, it is estimated there are approximately 486 parades with more than 10,000 attendees in the U.S. annually and that the total attendance for these parades is about 25.5 million. A summary of the case study and the scaling results for parades is provided in Exhibit 7-2.

No national data on parade revenue was available. As a result, average in-event spending per attendee at parades was estimated at $27.43 based on an average of fair and festival events data that was collected in the San Jose report that is also described in the Study Methodology section. Based on this estimate, the 25.5 million attendees at large parades annually spend approximately $699 million.

Fairs

National-level data documenting the frequency and attendance sizes of fairs does not exist. As a result, national estimates for these types of events were developed using data from the case studies of four cities conducted as part of this research. The only case study city that provided data on fairs was Columbia, which had 17 fair event days with more than10,000 attendees. This estimate was divided by four to represent the average number of fairs per case study city. Using this methodology, it is estimated there are approximately 1,955 fairs with more than 10,000 attendees in the U.S. annually. The average attendance at these fairs is assumed to be 12,500, the same as the Columbia average. Accordingly, fairs with more than 10,000 attendees are estimated to attract a total of 24.4 million attendees annually. A summary of the case study and the scaling results for fairs is provided in Exhibit 7-3.

These results also appear reasonable against two other estimates. The International Association of Fairs and Expositions notes that "Today, over 3,200 fairs are held in North America each year. They provide industrial exhibits, demonstrations and competition aimed at the advancement of livestock, horticulture and agriculture with special emphasis placed on educational activities such as 4-H, FFA and similar youth development programs."⁵⁵ Additionally, the International Association of Fairs and Expositions website includes a database of fairs that allows the user to list the fairs by state, including data on the name of the fair, the city, the state, and the dates of the fair. This state level data can be used to extrapolate national averages. For example, there are 22 fairs listed for the state of Indiana. If Indiana is a representative state, the database could be expected to list over 1,000 fairs in the U.S.

Revenue estimates for fairs were derived using the average in-event attendee spending figure of $27.43 for fairs and festivals developed from the San Jose study. Accordingly, it is estimated that about $670 million is spent by 24.4 million attendees at large fairs annually.

Festivals

There are no comprehensive data sources on festival revenue and attendance on a national level. National-level estimates were developed by scaling local results from the four case study cities. Using this methodology, there are an estimated 3,110 large festivals in the U.S. annually, which are estimated to have a total attendance of about 111 million. Columbia appeared to have an abnormally large number of festivals, so the results from that case study were reduced from 13 to 5 by scaling the results for El Paso based on population. A summary of the case study and the scaling results is provided in Exhibit 7-4.

The in-event spending estimate for festivals is based on the $27.43 estimate derived from the San Jose study. Total in-event spending for festivals is estimated to be about $3.0 billion.

Protests and Political Events

Frequency and attendance estimates for protests and political events across the U.S. are poorly documented. As a result, national-level estimates were developed by scaling local results in four case study cities conducted as part of this study. The case studies of Detroit and El Paso did not record any protests or political events. To smooth out the data for Very Large cities, data for this category was assumed to be the same as for Large cities. Data for Medium cities was smoothed out by substituting in Small city data.

Using this methodology, there are an estimated 499 large protests and political events in the U.S. annually. These events are estimated to have a total attendance of 14.1 million. Data for Medium cities were based on the Small city data, as no observations were recorded in the El Paso case study. A summary of the case study and the scaling results is provided in Exhibit 7-5.

Revenue estimates for protests and political events were derived using the in-event attendee spending average of $27.43 for fairs and festivals developed from the San Jose study. Accordingly, it is estimated that about $387 million is spent by 14.1 million attendees at large protests and political events annually.

Expositions and Shows

The most comprehensive list of conference centers in the U.S. is provided by Wikipedia.com, which lists 216 conference centers in an article titled "Convention Centers in the United States."⁵⁶ This number is close to the number of conference centers in the U.S. that are part of the International Association of Conference Centers.⁵⁷

To allocate the 216 conference centers to the four city size categorizes, it was assumed that Very Large and Large cities had one conference center each and that the remaining conference centers were in Medium and Small cities. No expositions or shows were observed in El Paso; whereas Columbia recorded 8 event days. In order to smooth out the data, the average of the El Paso and Columbia cities was applied to both Medium and Small city categories. Exhibit 7-6 provides a summary of the case study and the scaling results.

Revenue estimates for expositions and shows were derived from a report on the San Diego Convention Center.⁵⁸ This report estimated that the convention center had $33.8 million in revenue and about 984,000 visitors, for an average revenue of $34.44 per visitor. Therefore, it is estimated that 43.3 million visitors spend about $1.5 billion annually at expositions and shows across the U.S.

Economic and Fiscal Impacts

Measuring the economic and fiscal effects of large PSEs is essential to a better understanding of the role that these events play in the economic vitality of our cities and our nation.

The economic impact of spectator spending and revenue on the local economy can be estimated using a multiplier. In economics, the multiplier effect refers to the idea that an initial spending rise can lead to an even greater increase in local spending and income. In other words, an initial change in aggregate demand can cause a further change in aggregate output for the economy.

Assigning economic and fiscal impact multipliers to the 17 event type categories is difficult due to the lack of exact matches between this report's event categories and the event categories in secondary studies regarding economic and fiscal impact. For consistency, all the economic and fiscal impact multipliers in this report, except for expositions and shows, are based on an economic and impact assessment of events in San Jose.⁵⁹ The San Jose report has the advantage of covering multiple event types, using actual data collected from 10,000 actual event attendees, and defining economic impacts to include only outside-event spending that would not have occurred without the event. Multipliers in the San Jose study were assigned to the most closely-matching event type categories in this report. The economic and fiscal impact multipliers assigned to each event type in this report are listed in rows 17 and 19 of Exhibit 2-13. Since data by event type represent only in-event spending, the multipliers combine outside-event spending with indirect and induced multiplier effects.

Total economic and fiscal estimates were developed using data for the 17 event type categories discussed in earlier sections of this report for in-event revenue or spending-per-attendee estimates. Economic and fiscal impacts were estimated to be the product of an appropriate multiplier and either an average revenue per attendee estimate or an average spending per attendee estimate. The results of these estimates are provided in Exhibit 7-7. For example, professional football has a total of 19.4 million attendees annually. The average in-event revenue per attendee is $337. This results in $6.7 billion in revenue for the event category. The economic impact of this initial revenue is $22 billion after taking into account an economic multiplier of 3.246. The fiscal impact to local, state, and federal levels of government is estimated to be $521 million using a fiscal impact multiplier of 0.024.

National Congestion Estimates

According to the Urban Mobility Report (UMR), the total annual delay per traveler for all of the 437 urban areas analyzed is 38 hours.⁶⁰ The annual delay per traveler nationally in the PM peak period would be 21.2 hours. The percent of this delay attributable to PSEs is between four and eight percent of the total. Accordingly, the delay caused by planned special events results in between 0.8 and 1.7 hours of delay per year per traveler. The results of the congestion analysis for the 437 urban areas are summarized in Exhibit 7-8.

The next category analyzed in the UMR is wasted fuel per traveler, which is 26 gallons per year. Wasted fuel per year attributable to the PM peak period is 14.5 gallons, which translates to wasted fuel per year of between 0.6 and 1.2 gallons due to planned special events.

The travel delay as stated in the UMR for the 437 analyzed urban areas is 4.1 billion hours of total delay. The delay in the PM peak period was calculated as 2.3 billion hours. Planned special events, therefore, account for between 93.4 million hours and 186,824,000 hours of delay in the 437 urban areas analyzed in the UMR.

Planned special events account for between 4 percent and 8 percent of all PM peak period delay. The remaining delay is caused by a list of many other factors. The key aspect of this congestion is that the delay affects both attendees as well as non-attendees of the planned special events. The attendees experience delay while they are trying to enter the various venues and parking facilities for the event. However, this delay is almost expected, since attendees of planned special events realize that the planned special event will increase traffic and disrupt normal traffic patterns. .

Further analysis of the UMR reveals that the excess fuel consumed in the 437 urban areas due to congestion is 2.8 billion gallons per year. The excess fuel consumed in the PM peak period was calculated as approximately 1.6 billion gallons, which translates to between 63.9 million and 127.9 million gallons per year due to planned special events.

Finally, the congestion cost as determined by the UMR for the 437 urban areas analyzed as part of the report is $78 billion dollars. The congestion cost in the PM peak period would translate to $44 billion, and the congestion cost of planned special events would work out to between $1.7 billion and $3.5 billion per year.

A close look at the numbers reveals that planned special events have a significant effect on the traveling public throughout the country. A total cost of $3.5 billion for the 437 analyzed urban areas is substantial.