When Ticket Revenue Doesn’t Equal Attendance: What’s Opera Doc?

*THANKS TO AARON ANDERSEN FOR POINTING OUT A GLARING ERROR I MADE (see strikethrough texts below and my comment following Aaron’s)

I was reading a piece about the sharp decline of NASCAR ticket revenue and was intrigued. In NASCAR’s three publicly traded companies, all have seen a sharp decline over the years.

For example, at Daytona Beach, the International Speedway Corp.lost more than 40 percent of its ticket revenue, falling to $144 million” while the Charlotte Speedway Motorsports Inc.has lost more than a quarter of its admission revenue, falling to $130 million.” Dover Motorsports Inc. took the biggest hit “with admission revenue falling nearly 60 percent, to $13.6 million last year.

The piece gives various reasons for the decline in ticket revenue, and offers some solutions the different franchises are considering or actively doing, yet, this statement is interesting given what can amount to a loss of 57,000 (current capacity of Daytona Speedway is 147,000) butts in the seats:

Television ratings, which are roughly flat from 2010 to 2011, show that the sport remains popular, even if fewer fans come to the track.

While a live audience of 90,000 isn’t something to scoff at, a loss of roughly 40% of an audience would be considered alarming in the performing arts field (or any other industry for that matter). Note also that there are current plans to renovate the Daytona Speedway to bring the total capacity down to 101,000 (by 2016) from its current 147,000 limit–a 31% reduction at an estimated cost of $400 million. This will be its fourth major renovation and keep in mind that one of those reduced its capacity from 162,000 down to its current 147,000 level.

How Sports Attendance Figures Speak Lies

A piece in Forbes by Maury Brown opens by touting how much “attendance is one of – if not the – key revenue streams.” Obviously, as I’ve mentioned many times here at this blog, ticket or gate revenue constitutes a smaller proportion of total revenue earned by Sports franchises than we see in the performing arts (as low as 20% in the case of the NFL) so obviously the author isn’t talking solely about the low attendance revenue.  Or rather, he’s talking about how attendance helps to generate non-performance revenue.  As he states:

Have a bunch of sellouts, and it drives sponsorships. Come up short in the NFL, and you have television blackouts. The number reported in Major League Baseball determines factors as it pertains to revenue-sharing.

In other words, attendance is a big deal.

But, here’s the thing: since attendance has become such a key component for sports leagues, the actual idea that it shows how many people are actually at a game is a fantasy.

He proceeds to problematize the equivocation of Tickets sold as a reflection of attendance. That problematization might be interesting to look at in light of a Bizjournal piece from last year discussing what’s problematic about the purportedly good year the Chicago Lyric had last season. All of this is timely given the situation at the Met Opera and the impending labor negotiations.

Paid Attendance is not “attendance”

Brown opens up his critique with the above statement as a section header and is from where I got the title of my blogpost. On the face of it, it shouldn’t be a particularly controversial statement.  If someone buys a ticket, but doesn’t go to the event or performance, that is an example of “Paid Attendance” which isn’t “Attendance.” The organization made money from the ticket, but no one put their butt in that seat. It is possible to buy all tickets at an event without putting a single butt in the seat–a sold out show with no audience!

Alternatively, it is also possible to sell well past capacity without filling a seat.  For example, if 100% of the tickets sold were returned for resale, and any percentage of those were resold, and none of those ticket buyers showed up, then you could still have an empty hall while having resold up to 100% of returned tickets.

Filling a ballpark/concert hall past 100%

As Lazare states, there “was a time in the 1990s when Lyric consistently played to 103 percent of capacity for its season of operas.” He does this to contrast with the current (2012/2013 season) 83 percent of capacity sold which is down from 88 percent a year ago. As Greg Sandow states, “When subscribers couldn’t go to performances, they’d give the tickets back without asking for a refund. And the company would then resell those seats to single-ticket buyers.

Brown says, “If you polled 100 fans and asked them what a sellout of a sporting event meant to them, chances are all would say, ‘Every seat is the house was sold,” which is, strictly speaking, not true. As he explains:

In terms of Major League Baseball, sellout figures are often well below seating capacity.

Case in point, the Cleveland Indians announced last Friday that they reached a sellout for Saturday’s interleague game with the Reds, the first non-Opening Day sellout since May 24, 2008 when they played the Texas Rangers. But, when the numbers came in, you had to scratch your head.

Capacity for Progressive Field is 43,545. The announced attendance was 40,631, or 2,914 short of capacity. Sellout?

Brown gives several reasons to explain this. One reason could be to leverage for a new larger stadium, another could be for good PR (which could then lead to sponsorships for a consistently selling out club). Each club determines the threshold for capacity seating (not the actual seating), and it works that way for other Sports Franchises.

The NFL will show a sellout event to prevent a media blackout which is a rule instituted to encourage live attendance, and ultimately affect broadcast licensing revenue. Of course, by limiting the seating capacity in this Sport means that capacity is set for the season and further limiting isn’t possible.  In that case, the team owner or other sponsoring organizations or corporate partners can buy up the remaining tickets to prevent the blackout. Brown gives us another example with commentary:

A great case was last year when restaurateur Russell J. Salvatore made a last-minute purchase of 7,000 to 8,000 tickets so that the season finale for the Buffalo Bills would not be blacked out. Normally, corporations step in, or even the club itself, will purchase tickets, to reach the sellout number. But, clearly, Salvatore was a hero on the last day of the regular season for Bills fans.

None of this helps us understand actual attendance figures though we may have perfectly good information about ticket sales.

Going back to the Chicago Lyric, since we ‘re told the ticket sales number, for all we know the attendance could have been lower during the 103% selling capacity. We’re not given the returned ticket numbers: if anything more than 30% of tickets were returned (making these function, in effect, as small donations), then the 3% above 100% resells could constitute a lower number than the 83%–assuming all of the latter tickets were used. Since we’re given no information about the ratio of attendance to ticket sales we’re left with highly ambiguous figures which tells us nothing more than how many tickets were sold.

Interestingly, since the 103% figure comes from the 90s, we have a possible rationale for why ticket sales might have become more important to report than actual attendance.  As Brown states:

If you followed the business of sports, you’ll know that about 20 years ago, the move to count tickets sold, as opposed to turnstile clicks, became what is reported in the boxscores. The reason – wholly done for accounting purposes for the clubs – showed what matters to owners most: money.

Emphasis is mine and the period in question is analogous. It would be interesting to know if attendance figures rather than ticket sales were more often reported on before the 90s in the performing arts.

Conflating ticket sales with attendance

Given the above and how ticket sales are conflated with attendance figures, what happens when we look at industry wide ticket sales versus industry wide attendance? Fortunately we can take a look at how big a disparity actually exists between the two.

Greg Sandow, in the link I posted above, had his assistant speak to the research director of Opera American, Kate Place. In his own words at the blog states he states:

Back in September, I asked my invaluable assistant, Caroline Firman, to contact classical music service organizations in the US, and ask if they had information on ticket sales in the area of classical music they deal with. Kate Place, research director of Opera America, supplied a chart showing attendance from 1988 to 2012 at all the larger US opera companies, taken together:

Note the bolded (my emphasis) sections. Here is the chart,

Opera_America_1988_-_2012_Attendance__1_page_

and to be fair, Sandow emphases that this is just information from larger Opera companies.

These numbers aren’t complete. We don’t know how smaller companies are doing. Kate Place says that not enough of them report their data for her to feel that she can say how they’re doing as a group.

We don’t know (or at least aren’t given info about) whether the chart is about attendance figures or ticket sales. Unfortunately, we’re also not given a distribution table of the figures so we’re left to guess the figures give. While the chart does end with a figure that’s lower than the 1988 number, notice the variance in the figures throughout.  We’re not given an analysis of the variance, we can’t tell if the figures fall within variance norms but there is an obvious downward trend from 2001 to 2012.

I’m going use the number from 2001, which seems to be the high point of “attendance.” Let’s just arbitrarily set that number at 1.75 million. Fortunately, we do have other sources for data for 2001 about “attendance” (and attendance) or “ticket sales” (since we’re not sure what this chart is actually showing us). Interestingly, one of the sources we have is also gleaned from Opera America data.

The National Arts Index 2013 Report shows (pg. 68) opera attendance to be at 3.872 million in 2001 (it gives the figures from 2000-2011). Where does this number come from? Opera America! NAI also gives its caveat.

This refers only to main stage performances, so it certainly understates the total audience. Like symphony, theatre, dance, and other art forms, much opera activity is offered in educational and community settings to large audiences. Those audiences, however, are not systematically counted. More significantly, these figures do not include the very popular simulcasts of Metropolitan Opera performances, which have welcomed millions more viewers to the opera experience.

Note also that most smaller companies don’t report their data, so that isn’t included in this more robust number of attendance. It also should be noted that Opera America, on its Quick Facts page lists opera attendance during the 2009-2010 season at 6.7 million–well over the one million mark on the supplied graph above.  How much bigger would the attendance be if the smaller companies were included?

To be fair, the NAI Report also shows a decline: by 2011 the figure drops to 2.304 million.  Not nearly as much of a drop as we see from the chart of larger companies (1.75 million to 807,857), and the cut off is at 2011, but unless we see a drop to roughly 1.8 million (roughly half a million), which isn’t inconceivable, then we’re talking about a far less significant decline.  But decline it is–“but a decline in what?” is the question.

Let’s turn to the National Endownment for the Arts Survey of Public Participation. In the 2002 survey, which measures the preceeding year (2001), the adult population of 205.9 million had a 3.2% attendance rate.  Note, this is a randomized survey of self-reports, the latter condition of which is problematic for various reasons. That percentage, when converted from the adult population figure is roughly 6,588,800 (which is close to the average given in this Princeton Quick Fact sheet about the 2002 SPPA of 6.6 million attendance).

Note that this is just the attendance of the adult population–which means, much as the NAI pointed out “much opera activity is offered in educational and community settings to large audiences” is left out.  This is where we would find more children and adolescents in attendance. Also note that this gives some confirmation that the caveats given by the NAI about attendance seems to be somewhat true–that their number understates the total audience.

Again, to be fair, the NEA SPPA’s since they were first used in 1982 has shown a consistent decline in self-reported attendance to Opera events.  But how much of a decline–and a decline in what is the question. The numbers above don’t seem to differentiated between tickets sold versus actual attendance (butts in seats), or they are from self-reports extrapolated from a population sample.

Over 200 Opera Organizations have formed in the US since 2000

As most readers of this blog knows, I’ve been compiling a list (amongst many) of Opera Organizations formed since 2000.  The list currently has over 200 opera organizations listed (and may be found here) and I’m constantly coming across several groups each week since I’ve started the list.

Going back to the NAI, it states that:

Opera America, the national service organization for the opera field, conducts an annual Professional Opera Survey. The more than 100 opera companies that respond annually to this survey produce hundreds of productions and over 2,000 performances annually.

Since the previous section highlighted understated audience numbers, here is another side to consider.  If Opera America’s Professional Opera Survey includes a little over 100 self-reporting opera companies (mostly bigger organizations) then how few of these recently formed companies are actually included, and how much is attendance understated given the proliferation of newer companies in recent years?

Sure, we all understand that most of these smaller companies ARE small: chamber opera companies with small cast of singers and usually only piano accompaniment playing in small capacity venues. But some are Grand Opera companies–many of which formed in the dissolution of similar companies in particular regions. And many of these companies are doing far more performances per season than some of the big budget and established Grand Opera companies.

In other words, we have no idea what the aggregate audience is of these new companies and how they figure into the attendance data. Sure, many of the self-reports from the SPPA might reflect some of this attendance–we’ll know more once we see if these proliferating organizations is a growing trend or simply a cultural bubble. Note that the 2008 and 2012 SPPA both report that 2.1% of the adult population attended opera. While this number is down from the 3.2% of 2002, it might also indicate that the absolute number of attendance is growing while the proportion is remaining the same, and some of the reason for that could be the proliferation of these smaller companies which aren’t getting reported in the Opera America Survey which is showing decline in attendance of reporting companies.

Obviously, since we’re not sure if we’re seeing ticket sales numbers rather than actual attendance figures, things are up in the air. Ultimately, these are some of the problems and issues that can arise when we selectively look at data, and when that data isn’t representative of the field as a whole which constitutes sampling bias. Compound the two and we can eliminate all the nuances which can give us a much more accurate picture of the issues and problems, which will then lead to better solutions.

While live attendance is a useful measure to show how popular something is, it’s not the only measure. Live sports attendance, movie attendance, and pop music attendance has been declining–in some cases at an even greater rate than classical music attendance. But live attendance isn’t the only measure for popularity or health.  As the NASCAR example above states, television ratings show that the sport remains popular.  Even thought the NFL has an annual live attendance that is nearly a third smaller than live attendance at orchestras, it is viewed as popular because of participation through broadcast media.

The NEA SPPA shows us that Classical Music has the highest rate of online participation of any activity (with Latin Music a close second). These attendance figures don’t include attendance at livecasts of the LA Phil or Met Opera, nor do they included non-adult attendance which means the educational and public community event attendance (which tends to be very high) gets left out of the equation.

Opera America states on its Quick Facts page that of the opera companies, “Over half of these companies were established after 1970, and over 40% were established since 1980, indicating the growth of opera throughout North America in the last 40 years.” Since it “counts 122 company members in 43 states in its membership,” and since we know that since 2000, over 200 groups were formed, this trend of growth seems to have been happening for some time.

Whether or not this signifies the decline of big mainstream institutions is another question altogether, but using decline to characterize the field as a whole is probably premature, and in the end doesn’t help those of us who are involved in these newer companies and trying to build support, revenue streams, and audiences for them.

I know that I’ve taken a bit of a break from the blog recently, but it’s been a tremendously busy performing schedule for me and I wanted to fire this post off before I head up to Michigan to play in the pit with the Soo Opera (another group formed since 2000) in their production of Carmen. I just recently played six shows with the Thompson Street Opera Company (formed in 2011) last month–two new chamber operas, and have shared a bill on one event earlier in the year with Cincinnati’s NANOWorks Opera (formed in 2012) with my Klingon Music Project (launched in 2011) which has also produced its own Klingon Opera.

I hope to get back to regular blogging after next week, but till then, look at the whole Operatic picture. Until then, enjoy this classic.

Advertisements

3 thoughts on “When Ticket Revenue Doesn’t Equal Attendance: What’s Opera Doc?

  1. You nearly lost me long before I understood the point you were trying to make, for two reasons:

    1. You complained that Sandow didn’t include a specific data table and ANOVA, yet you couldn’t be bothered to notice the data scale on the left side of the graph. If you’ll refer to it, you’ll see that the high point is somewhere between 1.1 million and 1.2 million, not 1.75 million. This leads you to erroneous, perhaps egregious, overstatement of the audience decline for group 1 operas. And, when you complain about what’s not there while failing to pay attention to what clearly is, you undermine yourself.

    2, and more important. You started the piece by highlighting an apples vs oranges problem with attendance vs sales, then you started pulling in even more disparate data from NAI and SPPA. You added pineapples and kumquats to the apples and oranges.

    The basic tone here comes across as, “We can’t believe anything anybody says about the popularity of opera, so let’s assume more people like it than anybody says.” You’re certainly not going to find any definitive answer to how many people attended an opera performance in a given year (or a theater performance or a classical music performance or whatever). There will always be a lot of organizations that aren’t counting, or that aren’t counting in the same way. Everybody knows this. What’s far more important is to understand what the data you’re looking at include or excludes, and look for longitudinal consistency in what is included and excluded from year to year. That’s the best you’re going to be able to do, really.

    Several times, you’ve made the very important point that even if large institutional arts organizations are struggling, that’s not the same as the art form itself struggling. That’s important to understand. But this post didn’t help you make the point.

    Like

    1. Thanks for your comment, Aaron–just a few quick comments as I’m only in for a few more hours and need to get some sleep.

      “you’ll see that the high point is somewhere between 1.1 million and 1.2 million, not 1.75 million.”

      Yes, you’re correct–that was a typo. I meant “1.175” as the chart looks closer to three quarters to the way to 1.2 million. Thanks for pointing that out. And I obvious repeated that number in my other calculation as you noted–good call!

      What’s far more important is to understand what the data you’re looking at include or excludes, and look for longitudinal consistency in what is included and excluded from year to year. That’s the best you’re going to be able to do, really.

      Right–which is why I did note that all three data sets are actually showing decline longitudinally (ignoring the last two SPPAs as a possible outlier), as well as the few examples of problems I mentioned in the three sets (self-reporting bias issue, not randomized samples, not including children or non-live performance attendance, not including attendance figures from many smaller companies). The latter three could be consistent with the idea that attendance is understated.

      Several times, you’ve made the very important point that even if large institutional arts organizations are struggling, that’s not the same as the art form itself struggling. That’s important to understand. But this post didn’t help you make the point.

      Yes, what you’re referring to is the underlying important point that I’m implicitly stating, but what I really wanted to highlight is that we could also take the body of evidence in the data sets (as woefully incomplete as they are) as showing that the art form might simply be growing in ways that the data sets can’t accurately measure since they don’t exactly show us that there’s a decline (whatever the hell a “decline” is in first place) in the art form.

      Again—thanks for pointing out the error, that does make a big difference!

      Like

    2. I thought about it a bit more and I probably didn’t make the point I said I wanted to make at all–I had struggled with cutting down this post a lot as I originally wanted to talk about what we mean by popularity and how that relates to how we view the arts in general. I cut most of that out, but the tone of the post still has echoes of it which do seem to distract from the other point I wanted to make re: how we look at figures and conflate ticket sales with attendance (thus giving us ammunition for whether or not something is popular).

      Thanks for pointing out what was still a strong point in the piece (and one I wasn’t even focusing on in the first place)–I probably should have focused even more in the piece. Really do appreciate your criticism!

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s