« Does Santa Clause reduce the rate of savings? | Main | Accounting identities »
Is "media bias" just good business?
Austan Goolsbee writes (no permalink yet) of Matthew Gentzkow and Jesse M. Shapiro, and their paper “What Drives Media Slant? Evidence From U.S. Daily Newspapers”. The non-gated version is here.
In essence the authors measure how much newspapers use key partisan phrases like "personal accounts" for social security privatization and compare the vocabularies of those newspapers to the vocabularies of partisan politicians. The political slant of newspapers is then matched to campaign contributions in the zip codes those newspapers serve. Shapiro sums up the result:
The data suggest that newspapers are targeting their political slant to their customers’ demand and choosing the amount of slant that will maximize their sales.
It also turns out that the political views of the paper's owner have no effect on the slant of the paper.
Here is my previous post on wunderkind Jesse Shapiro. He remains the best Youngling out there. Here is an earlier piece I wrote on media bias.
Posted by Tyler Cowen on December 7, 2006 at 08:13 AM in Economics | Permalink
Comments
I'm not convinced. One of the main examples the author uses is the dichotomy between "death tax" and "estate tax". Shapiro says that using the phrase "estate tax" is Democratic in that it is used more by Democrats than by Republicans--fair enough. However, it is hardly slant; it is an accurate description of the tax. The term "death tax" is slant, while "estate tax" is neutral. The author even notes:
"Republicans put a high level of importance on the death/estate tax language--they had to work hard to get members to act in unison, including training members to say "death tax"... Estate tax sounds like it only hits the wealthy but "death tax" sounds like it hits everyone" (p.11)
Yet use of the phrase "estate tax" is deemed leftward slant.
The same is true for other tax related phrases. Using charged phrases like "tax relief" is deemed a slant equal to referring to actual spending cuts. Also, since when is "veterans health care" a Democratic phrase?
Further, ownership slant is perhaps measured better by the type and number of stories run on a subject (skepticism of the Iraq war, for instance) and how a story is cut by editors. Tom Ricks has seemed to suggest that his stories were sliced and diced by his editors for the past couple of years.
Posted by: eriks at Dec 7, 2006 9:52:52 AM
Good for them and their delusions, not good for the country.
Posted by: Sandy P at Dec 7, 2006 10:44:44 AM
Richard Posner wrote a great article almost a year and a half ago about this.
http://www.nytimes.com/2005/07/31/books/review/31POSNER.html
Posted by: Omar at Dec 7, 2006 10:46:59 AM
A lot of this would make sense. I think a lot of the right-wing uproar about the biased media stems from the past when there were much fewer media outlets, so there was not enough competition to force them to appeal to right-wingers. Fox News showed how profitable it could be to exploit that niche and I think their competitors have caught on, but the most vocal right-wingers seem to imagine that they are still living in the past.
Posted by: TGGP at Dec 7, 2006 10:54:16 AM
I always just assumed that everybody thought this already.
Posted by: josh at Dec 7, 2006 11:26:46 AM
This is a very good paper. But even if the result is right (and I certainly have no reason to believe it isn't), it doesn't mean that ideological persuasion in the media is not important. First, it may be the case that outlets don't deviate much from the profit-maximizing content, but the small deviations that they do make have a lot of persuasive power (editorial page electoral endorsements have been cited as a possible example). Another possibility is that people choose the outlets they do because they want to be persuaded. For example, suppose you have an image of yourself as a conservative. You might choose to watch Fox News as a form of reinforcement; to be a better, truer, more complete conservative. This is consistent with DellaVigna and Kaplan research that shows a *causative* effect of Fox News on voting, Republican party donations, and even military enlistment. This is completely consistent with the Gentzkow & Shapiro result, but still places a great deal of persuasive power in the hands of the owners of Fox News, who have a very great influence over what form that reinforcement will take and over what will be defined as a proper conservative position.
Posted by: David J. Balan at Dec 7, 2006 11:37:28 AM
professor goolsbee unfortunately doesn't go far enough with his commentary (perhpaps he's censored himself given the liberal forum, but would be obvious extension from his fabulous Econ of TMT class at the GSB) - this is prime evidence that media ownership regulation is outdated and needs to be repealed.
Posted by: ptkelly at Dec 7, 2006 3:15:39 PM
I would think that the slant would have more of an endogenous component to it, rather than just some owner/editor deciding the slant to maximize profits. If a paper starts out a little slanted relative to its competitors, it will be bought by those who typically share that slant themselves. On the paper's side, slanted editors and journalists will be attracted to the paper because it may feel less constraining then say if a coulter were writing for the NYTimes. To suggest that the papers are actively patrolling the slant inside their own paper while watching thousands of little markets to pick the optimal slant seems a little silly. I guess I could be missing the whole point, and this is just another example of how markets provide optimality without any direction, which goes to show the confusing language economists use personifying markets as active optimizers. But then again, if this is the case, whats the big story? That people buy newspapers that confirm their worldview? Slow news day i guess.
Posted by: Andrew at Dec 7, 2006 4:41:35 PM
How does this research square w/Karlan et. al (and actual randomized control experiment) that says the difference between receiving the Wash Post/Times was worth about 8 points in the last gubernatorial elections in VA???
Posted by: Dan K at Dec 7, 2006 5:06:32 PM
Probably good business. Probably not just good business.
Posted by: bk at Dec 8, 2006 9:42:26 AM
It never made sense to me that any true conservative could believe a free market was biased. A much simpler explanation is that their perception of media bias was actually due to the hostile media effect.
Posted by: fling93 at Dec 8, 2006 9:09:02 PM
So the reason why newspapers tend to tell things from a liberal point of view is because more liberals buy newspapers.
I guess that makes sense, although I can't help but wonder how much of this is a chicken and egg thing.
Posted by: Korgmeister at Dec 9, 2006 3:25:44 AM
The first comment was along what I noticed also.
They seem to be looking for Right-wing bias. If they go out looking for more of one than another, they'll find more of it to be sure, while bypassing Left wing slant altogether in many cases.
For anecdotal evidence, the Arizona Republic is largely out of step with their local demographics, but without a direct threat to their newspaper hegemony, they continue to be overtly Left-wing in their presentation. And the average citizen, if they buy anything, continue to buy the paper for lack of anything else to buy. (Yes, we are the state that voted for a very liberal female governor that most are quietly convinced is a lesbian, while simultaneously voting for a pile of ballot props that were deemed anti-immigrant by the Left i.e. English as an official language, and a few others.)
Posted by: Ray G at Dec 9, 2006 2:43:58 PM
The most common indicator that comes to my mind is the usage of the very words conservative and liberal.
The slightest nod to the Right by an individual, organization, or whomever is enough to forever be deemed by the media as "conservative." Conversely, it takes quite a radical stand to be openly identified as a "liberal" or Left wing by the media.
Also, Republicans are always identified as either conservative, or moderate, though I can't recall any widespread practice of identifying Democrats as moderates to distinguish them from their more liberal comrades.
Posted by: Ray G at Dec 9, 2006 3:15:19 PM
数控机床数控机床。。数控改造。。
数控机床。。
数控机床改造。。
数控机床仿真软件数控机床改造。。
数控机床。。
数控机床。。
数控机床。。
数控车床数控车床。。
数控车床。。
华中数控车床。。
大连数控车床。。
无锡数控车床山东数控车床。。
南京数控车床。。
数控车床。。
数控车床。。
广州数控车床数控车床。。
广州数控车床。。
华中数控车床。。
数控车床。。
数控车床北京数控车床。。
数控车床简介。。
教学车床。。
教学数控车床。。
数控铣床南通数控铣床。。
数控铣床。。
数控铣床。。
数控铣床。。
教学数控铣床教学铣床。。
数控机床实训设备。。
加工中心。。
加工中心。。
加工中心加工中心。。
微型加工中心。。
数控加工中心。。
加工中心。。
加工中心加工中心。。
加工中心。。
上海加工中心。。
加工中心
Posted by: 南京北春 at Aug 21, 2007 3:16:21 AM
