newspapers. In particular, newspapers increase subscription prices, sacrifice circulation,
and set lower ad prices (see
Seamans and Zhu, 2014).
After television advertising, Internet advertising has become the most important
medium in terms of ad revenues, at least in the US. As
Figure 10.2 illustrates, in 2012
advertising revenues for Internet media were close to 37 billion US$. As in other media,
advertising can play different roles for advertisers. It can inform about product availabil-
ity, price, and product characteristics; it may allow consumers to draw inferences about
product characteristics (advertising as a signal:
Milgrom and Roberts, 1986; Nelson,
1974
); it may change consumer preferences to the benefit of the advertiser and thus
be persuasive; or it may serve as a complement to the product (
Becker and Murphy,
1993
).
10
We return to these views on advertising when discussing different Internet
advertising formats.
11
Figure 10.3, in turn, shows the rapid increase in advertising spending on Internet
media, with an annual growth rate of about 20% over the last 10 years. Only at the height
of the financial crisis in the US from 2008 to 2009 can one observe a small dip in ad rev-
enues. This contrasts with the development of newspaper advertising revenues. In 2000, ad
revenues stood at around 49 billion US$; in 2012, they totaled less than 20 billion US$.
12
39.6
36.6
32.5
22.8
19.4
16.1
7.5
0.8
0.7
0
5
10
15
20
25
30
35
40
45
Figure 10.2 Advertising revenue market share by media—US$ billions in 2012. Source: PwC (2013).
10
Bagwell (2007) provides an excellent survey of the economics of advertising.
11
See also Chapter 4 in this volume.
12
There is compelling evidence that Internet advertising can be effective in stimulating purchases. This does
not hold only for online purchases, but also for offline ones. For example,
Lewis and Reiley (2014) find in
a controlled field experiment with a large retail store that more than 90% of purchase increases occur in
brick-and-mortar stores. Similarly,
Johnson et al. (2014) demonstrate that repeated exposure to online ads
was highly profitable for the retail store whose data were analyzed.
455
The Economics of Internet Media
Internet advertising can take different forms.
13
Here, we mention the most common:
search advertising accounts for a large fraction of total Internet ad revenues. To the extent
that it makes consumers aware of certain offerings, this type of advertising can be seen as
directly informative. Consumers may also consider it to be indirectly informative about
the quality of a match through a signaling role of the ad—e.g., if an offering appears at the
top of the page of sponsored search results.
The dominant platform for search advertising is Google; in the US it first took the top
position in 2003, relegating Yahoo to second place. In 2013, around two-thirds of all
search requests on general search engines were made on Google, while 18% were made
on Microsoft sites (using Bing) and 11% on Yahoo (according to comScore qSearch). In
Europe, Google is even more dominant, accounting for more than 90% of searches in
2013. Google is stronger yet in terms of revenues. An explanation for this finding is that
the larger platform is more attractive to advertisers.
The second big category of ad revenues is display advertising. Display advertising can be
contextual, as it can be linked to particular keywords or phrases. We can also refer to such
a strategy as tailoring. In addition, display advertising on the Internet can be personalized
if the advertising platform has knowledge about consumer characteristics. This then leads
7.3
9.6
12.5
16.9
21.2
23.4
22.7
26.0
31.7
36.6
0
5
10
15
20
25
30
35
40
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Figure 10.3 Internet ad revenues 2003–2012—US$ billions. Source: PwC (2013).
13
We refer to Evans (2008) for a detailed explanation of search and display advertising. See Goldfarb (2014)
for a discussion of the different forms of online advertising.
456
Handbook of Media Economics
to targeted advertising on the Internet,
14
which improves the match between the adver-
tised product and the consumer. In general, such advertising can be informative, persua-
sive, or complementary.
The third major category of advertising revenues used to be classified ads; however,
revenues have been decreasing over the last years in the US. In the early 2000s, classified
ads moved quickly from newspapers to Internet platforms (e.g., Craiglist in the US)—one
reason for the fast decline of advertising revenues at many newspapers. Classified ads on
the Internet allow users to apply individual searches rather than using a predetermined
classification scheme. Otherwise, the economics of classified ads did not change because
of the move to the Internet. One would often consider this advertising format informa-
tive, as it makes consumers aware of an offering.
Additional formats, which are listed in
Figure 10.4, include mobile advertising and
digital video advertising. Mobile advertising, which saw a quick increase between
2010 and 2012, has the potential to add another tailoring dimension: the displayed ad
may depend on a consumer being physically close to a particular location at which a prod-
uct or service is available. Advertisers’ hope is that mobile advertising is a means to gen-
erate immediate purchases. Here, advertising may play mainly an informative role, as it
makes consumers aware of a product or service that they may be interested in at a par-
ticular location. Digital video advertising is akin to television advertising, with the impor-
tant difference that it allows for tailoring and targeting (we note that television advertising
also allows for some tailoring). This format may be more attractive for advertising that is
persuasive or serves as a complement.
Figure 10.4 reports advertising revenues according
to the ad format for the US in 2012.
Media platforms on the Internet offer the possibility of measuring the impact of
advertising—by counting the number of clicks an ad generates, for example. This has
opened up the possibility of using a pricing model that is different from simply counting
the number of impressions, as is done traditionally in media. The measure for the latter is
cost per mille (CPM—i.e., the cost per 1000 impressions). If the ad price depends,
instead, on the number of clicks (or possibly even the number of purchases), the tradi-
tional pricing model is replaced by one that is performance-based. As
Figure 10.5
illustrates, performance-based ad pricing has been gaining the upper hand on the Inter-
net, accounting for close to two-thirds of all Internet ad revenues in 2012. Ad revenues
accruing to Google, through sponsored search results, make up the majority of these
revenues.
Internet advertising may affect not only own-product sales. Several researchers have
conducted natural experiments on Internet advertising to analyze spillovers of advertising
campaign on competitors. For example,
Lewis and Nguyen (2014) used display
14
Athey and Gans (2010) make the distinction between tailored and targeted advertising. See Sections 10.5
and 10.6
for more details.
457
The Economics of Internet Media
advertising on Yahoo!’s front page. On some days, Yahoo! sells ads on its front page as
ad-splits, which entail two display ads being alternately shown throughout the day, one
every even second and the other every odd second. This allows naturally for exogenous
variation because there is no systematic difference between even-second and odd-second
visitors, thereby providing a test group (visitors exposed to the target ad) and a control
group (visitors exposed to the other ad-split ad).
Lewis and Nguyen (2014) find that dis-
play ads caused an increase in searches not only for the advertised brand (35–40%) but also
for competitors’ brands (1–6%). The effect on searches by brand can be up to 23% for
16.8
7.7
2.6
3.3
2.2
1.8
0.7
1.1
0.0
0
2
4
6
8
10
12
14
16
18
Figure 10.4 Ad revenues according to formats—US$ billions in 2012. Source: PwC (2013).
46%
48%
45%
39%
37%
33%
31%
32%
41%
47%
51%
57%
59%
62%
65%
66%
13%
5%
4% 4% 4%
5%
4%
2%
0%
10%
20%
30%
40%
50%
60%
70%
2005 2006 2007 2008 2009 2010 2011 2012
% of total revenue
CPM
Performance
Hybrid
Figure 10.5 Internet ad revenues by pricing model. Source: PwC (2013).
458
Handbook of Media Economics
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.