which are not relevant to this study.
In addition to these exclusions, where data for some publishers was unavailable or inaccurate, these publishers were
not included.
The size of the sample varied across years due to missing values for some publishers. Different sample sizes did not
affect the reliability of the estimation under the fixed effects methodology which is described below.
Table 4: Sample size by year
Sample size by year
2011
2012
2013
Number of publishers
56
63
48
6.1.2 Data
The results presented in this report are based on free publicly available data, as well as data obtained from
comScore, Nielsen, Mint Global, and the OJD in Spain. The table below describes the variables included in the
analysis.
Table 5: Data summary
Data Description Source Revenue • Total revenues of publishers in the sample in USD.
source
Revenue
Mint Global,
• Total revenues are composed of both print and online revenues.
publishers’
• For cases where a publisher is a holding company for various businesses, revenues have been
annual
disaggregated for the individual publication or group of publications using annual report data.
reports,
• Revenues in local currencies were converted into USD using the average exchange rate for the
Oanda
given year.
GDP per capita
World Bank
• A metric of the value of a country’s economy per person
living in the country.
• GDP is defined as the total value added by producers in a country.
Circulation
Country
• The average number of print copies distributed per issue.
circulation
bureaus
Paywall
Publisher
• Represents whether a publisher had either a hard or soft paywall in place.
websites
• This variable takes a 1 if a paywall was in place for the period considered and a 0 otherwise.
Advertising spending
Nielsen,
• Represents the dollar amount that publishers spent on advertising their own products and
Oanda
variables on log(revenue) across countries. Due to data limitations, it was not feasible to estimate the country-specific
effects with the chosen specification.
The estimates of the coefficients are detailed in the table below.
Table 6: Econometric results
Fixed effects model
Number of observations: 169
Number of publishers: 66
Time period: 2011 - 2013
Independent variables
Coefficients
log(total traffic)
0.0641**
2012
-0.0583**
2013
-0.0651
log(GDP per capita)
3.495***
log(circulation)
0.178
paywall
-0.0491
log(advertising spending)
0.00135
Constant
-20.93*
*** significant at 1%
** significant at 5%
* significant at 10%
Standard errors were clustered on publishers
The estimated coefficient of log(total traffic) suggests that holding all else equal, a 1% increase in total traffic to a
news site is estimated to increase revenue by 0.0641%.
To estimate the dollar value of traffic to newspaper publishersʼ websites, the coefficient was used to assess the
impact of a 100% decrease in total traffic. The results imply that a 100% decrease in total traffic would lead to a
6.41% decrease in total revenue.
The coefficient of the variable log (GDP per capita) implies that the variable has a positive impact on publishersʼ
revenues. As the citizens of a country grow wealthier, publishers may attain higher revenues. The countries examined
also exhibit different GDP per capita patterns over the period.
Negative coefficients on the variables 2012 and 2013 indicate a declining trend in newspaper publishersʼ revenues
over time. This finding is consistent with the general market trends discussed in the main body of the report.
The time period of three years does not allow for significant year-on-year variations in some of the other key variables
estimated by the fixed effects model. As a result, the properties of the fixed effects model do not identify a significant
impact of circulation and paywall on revenue. However, an alternative approach undertaken using a random effects
model does capture a positive and significant effect of circulation, even when controlling for other firm-specific factors,
while also maintaining a similar, significant coefficient on log (total traffic).
Given the advantages of estimating a fixed effects model in this study, the random effects model was not used in this
report.2
Finally, the model has not identified a statistically significant relationship between advertising spending, as measured
by log(advertising spending), and revenues for these publishers in the sample over this time period for similar
reasons as for circulation.
6.1.5 Robustness checks
To assess the sensitivity of the estimates, additional specifications were run to test how the coefficients of interest
varied if other variables were excluded. These specification tests did not produce material differences in the
coefficient on log(total traffic) and did not affect its significance.
Recent literature by Cozzolino and Giarratana (2014) suggests that endogeneity may have been introduced with the
inclusion of both circulation and total traffic variables in the model3.
Additional specifications omitting circulation as well as employing a two-stage least squares estimation using the
instruments proposed by Cozzolino and Giarratana have been estimated. Differences between the coefficients of
these specifications and the original specification were not material.
6.1.6 Calculating the value of a visit
The average value was derived from the estimated revenue impact on the publishers in the sample and the traffic
they received as follows:
Estimates revenue impact / total traffic = average value of a visit
The estimated value of a visit ranges between €0.04 and €0.08. The range represents the average value across the
four markets for the years covered in the sample.
6.2 Relevant literature
The main sections of the report and the appendix reference a number of sources that provide evidence for statistics
and arguments made in the analysis. In addition, the study has consulted and taken into consideration other research
on the subject of newspaper publishers, website traffic, and market performance. Summaries of the literature are
presented below.
The impact of news aggregators on internet news consumption Athey & Mobius (2012) analyse the impact of news
aggregators on the quantity and composition of news in France.4
Using a case study analysis, where Google News added local content to usersʼ home pages who chose to enter their
location, they find that the inclusion of local content by Google News had mixed effects on local new sites. It
increased traffic, especially in the short run, but it also increased the reliance of users on Google News for their
2 For more details on panel data models please refer to Wooldridge (2010), “Econometric Analysis of
Cross Section and Panel Data”
3 Cozzolino and Giarratana (2014), “Mechanisms of Value Creation in Platform Markets”
4 Athey,S. and Mobius,M. (2012), ʻThe impact of news aggregators on internet news consumption: The case of
localisationʼ , mimeo, Harvard University.
choice of news, and increased the dispersion of user attention across outlets.
Furthermore, they find that the adoption of Google News leads to greater consumption of local news, both
unconditionally (by more than 26%) and conditional on Google News page views. They find a 5% increase in direct
navigation to local outlets (bypassing Google News altogether, presumably because the user had learned that they
like the outlet and actively chooses it in the future), and a 13% increase in clicks on local outlets from the Google
News home page. However, over time, incremental local news consumption is derived primarily from increased use
of Google News.
The impact of online news advertising on print advertising Sridhar and Sriram (2014) analyse how firms choose to
transfer their budgets over time between online and print newspaper publishing. Using monthly advertising data for a
large US newspaper over the period 2007-2011, they carry out an empirical study to determine whether online
advertising is substituting print advertising revenues.5
They find that, whilst online traffic is increasing, online advertising prices remain low. They report that 4-9% of the
decline in print advertising revenues was due to substitution from the transition to online media, with most of the
decline being attributed to the substitution of advertisers to alternative media options other than newspapers.
Media, aggregators and the link economy Dellarocas, Katona, and Rand (2012) develop a theoretical model to
determine how hyperlinking affects the incentives of news providers to produce quality content, rather than link to
third party content, and the resulting impacts on the profits and content quality of both news providers and
aggregators.6
They find that the Internet has been disruptive in breaking up geographical monopolies, with all content competing for
online readers. Linking allows similar sites to coordinate content production in ways that increase their joint profits
and quality, thus benefiting consumers.
The main benefit of aggregators to content creators comes from traffic expansion. Assuming content aggregators
form links to the best available content, their presence makes it easier for consumers to access good content, and
increases the attractiveness of the entire content ecosystem. However, the presence of aggregators incurs costs that
need to be considered, such as the appropriation of attention and revenues to news content providers. Their net
effect is positive for content creators only if the traffic expansion they induce is sufficient to offset the loss of attention
and advertising revenue.
5 Sridhar, S. and Sriram, S. (2013), ʻIs online newspaper advertising cannibalizing print advertising?ʼ, University of
Michigan.
6 Dellarocas, C., Katona, Z., and Rand, W. (2013), ʻMedia, aggregators and the link economy: Strategic hyperlink
formation in content networksʼ in Management Science 59:10, 2360-2379.