Elena Argentesi and Lapo Filistrucchi, "Estimating Market Power in a
Two-Sided Market: The Case of Newspapers", Journal of Applied
Econometrics, Vol. 22, No. 7, 2007, pp. 1247-1266.

The data used in this paper come from several sources. Information on
newspaper circulation comes from ADS, the Italian audit bureau of
circulations (www.adsnotizie.it). Survey data on readers' characteristics
come from Audipress (www.audipress.it).  Data on newspaper advertising
revenues and aggregate costs are instead published by FIEG, the publishers'
association (www.fieg.it).  All these databases are publicly accessible on
paper, and only data for recent years are downloadable from the
above-mentioned websites. Other relevant information, such as cover prices,
advertising prices, the dates of introduction of the supplements, the dates
of introduction of the online versions, and the dates in which editors
changed, were kindly provided by newspaper publishers. Here is a description
of the variables used in the estimation. We distinguish between the data
used for the estimation of readers' demand from those used for the
estimation of advertising demand since the two have different sample period
and frequency.

The data on readers' demand is contained in the files readersjae.dta and
readersjae.txt. These contain the same information. The former is a Stata
dataset, and the latter is a comma-delimited text file. There are 8736
observations.

The data on advertising demand is contained in the files advertisersjae.dta
and advertisersjae.txt.  The former is a Stata dataset, and the latter is a
comma-delimited text file. There are 48 observations.

The two text files, which are ASCII files in DOS format, are zipped in the
file af-data-txt.zip. The two Stata files are zipped in the file
af-data-dta.zip.

1) Data on readers' demand (monthly obs. from 1976 to 2001 included)

id		newspaper

day		day of the week

month		month

year		year

mkts 		market share of daily printed copies in each day of the week
		of each month for each of the four newspapers (computed with
		respect to the total population above the age of 14)

outmkts		market share of the outside good

rpr		real cover price in ITL (computed with respect to the
		consumer price index base 1995 "cpi")

wmkts		"within" market share (market share within our sample of
		newspapers)

dsuppl		dummy for the introduction of a weekly magazine of general
		information (this variable is always equal to 1 after the
		month in which the supplement was first issued)

dsupplday	dummy for the day of the week in which the generalist
		magazine is issued

ddon		dummy for the introduction of a weekly women's magazine (this
		variable is always equal to 1 after the month in which the
		supplement was first issued)

ddonday	       	dummy for the day of the week in which the women's magazine
		is issued

dgio		dummy for the games with prizes which could be played by
		buying a copy of the newspaper (equal to 1 for the whole
		duration of the games) 

dwebsite1	dummy for the presence of a website of the newspaper

ncr		number of local editions

dcod1...dcod7	dummies for the editors of "Il Corriere della Sera"

dred1,dred2	dummies for the editors of "La Repubblica"

dstd1...dstd7	dummies for the editors of "La Stampa"

dgid1...dgid3	dummies for the editors of "Il Giornale"

nothgio		sum of the characteristic "dgio" for the other newspapers
		(instrument)

nothsupplday	sum of the characteristic "dsupplday" for the other
		newspapers (instrument)

nothdonday	sum of the characteristic "ddonday" for the other newspapers
		(instrument)

t		time trend

cpi 		consumer price index (base 1995)


2) Data on advertising demand (yearly obs. from 1992 to 2003 included)

id		newspaper

year		year

modmkts	        market share of average daily advertising slots in each year
		for each of the four newspapers (computed with respect to
		the potential market of all daily publications, "daimod")

outmodmkts	market share of average daily advertising slots in each year
		for the outside good

daimod          average daily advertising slots in each year for all daily
		publications

rpr		real price of an advertising slot in ITL (computed with
		respect to the consumer price index base 1995 "cpi")

rprout		real price of an advertising slot for the outside good

prints		average daily printed copies for each newspaper

male		percentage of male readers

avage		average readers' age

free		average daily number of complimentary copies for each
		newspaper (instrument)

summale		sum of the characteristic "male" for the other newspapers
		(instrument)

sumavage	sum of the characteristic "avage" for the other newspapers
		(instrument)

cpi 		consumer price index (base average of 1995)