The late edition

THE John Bates Clark medal is awarded every year to an American economist under the age of 40. Past winners include such grandees as Milton Friedman and Paul Krugman. This year the American Economic Association honoured Matthew Gentzkow, an economist at the University of Chicago, for his work on a subject of particular interest to this newspaper: the volatile economics of the news business.

As Mr Gentzkow points out in recent research, newspapers’ woes are not due entirely to readers’ defection to free alternatives online. Time spent reading newspapers did indeed fall by half between 1980 and 2012, but most of the drop came before 2000, while the web was in its infancy. From 2008 to 2012, as time spent on the web as a whole soared, time spent reading newspapers fell much more slowly. Enchanting cat videos, in short, do not seem to have crowded out much news consumption.

Rather, it is a plunge in advertising that has hit newspapers hardest. Their ad revenue, adjusted for inflation, is back to the level of 1953. From 2008 to 2012 the revenue for every hour readers spent perusing a printed newspaper fell by almost half, as the web provided advertisers with an exploding supply of alternatives. The news is not all bad, Mr Gentzkow reckons: online ads can be better targeted to readers and are increasingly valuable to advertisers. Ad revenue per hour spent looking at newspapers’ online editions actually rose from 2008 to 2012. But readers spend vastly less time with online news stories than with their print counterparts. The web, in other words, is squeezing revenues where attention spans have proved durable and is boosting them where attention is fleeting.

Early in the web era publications needed to decide whether online news would cannibalise their existing business or complement it—and could therefore be given away. Many papers bet that online consumption might whet readers’ appetites for news, much as free airtime on the radio boosts rather than undermines album sales. Some enjoyed a brief rise in profits as a result, as fast-growing online-ad income supplemented earnings from print.

Yet in an analysis in 2007 of an earlier survey of the news market of Washington, DC, Mr Gentzkow found that this symbiosis was superficial. Some voracious readers reacted to free online news by consuming much more news overall. This helped total reading to rise, obscuring a shift among the majority of readers away from paid products towards free ones, Mr Gentzkow found. Had more newspapers charged for online content (as growing numbers, including The Economist, have since opted to do), print readership and profits might have been higher.

All the news that fits your beliefs

The travails of old media would trouble few outside the industry, but for the fact that the news business nurtures an informed citizenry, a public good of some value. Mr Gentzkow’s work also explores the effects of the upheaval in the industry on the marketplace for ideas. In particular, he examines the fear that both the provision and the consumption of news in America are becoming more partisan, with biased news outlets devoting little time in their programming to alternative points of view and biased consumers not bothering to seek them out elsewhere.

It is perfectly rational, argue Mr Gentzkow and a frequent collaborator, Jesse Shapiro, also of the University of Chicago, for readers to mistrust publications that challenge their prior convictions (that Elvis Presley is dead, for instance). Since readers seek out news that confirms their own views, publications have an incentive to specialise in an ideological niche, so as to cultivate a loyal audience. And that, it turns out, is what they do: slant in local papers tends to track local political beliefs rather than the ideological predisposition of newspaper owners or editors, the pair found in a paper published in 2010.

In competitive markets, however, this specialisation breeds a diverse marketplace, with lots of papers hewing to a variety of different world views. That, the authors reason, raises the odds that outright dishonesty in reporting will be exposed, and places a check on how distorted news can become. In a paper published in 2006 they found that more competitive media markets were less biased in their coverage of the 2000 presidential election, as measured by the share of time devoted to each candidate.

A diverse market would still be a worry, however, if it prompted readers to cocoon themselves in ideological “echo chambers”. Such self-segregation could impede transmission of accurate information and throw a wrench in the already creaky mechanics of democracy. Yet in a 2011 paper, Messrs Gentzkow and Shapiro conclude that these fears are overstated. The authors define a publication’s slant by examining the share of readers reporting a particular ideological leaning (“liberal” or “conservative”). They then examine detailed browsing data to see, for example, whether visitors to a given “conservative” site tend to stick with right-leaning news sources or venture more widely.

Online news is more segregated than television news or local newspapers, the authors find, but less so than national newspapers (and much less so than personal interactions among friends and family). Most consumers turn to mainstream sources for at least some of their news—on the weather, sports or business, say. Many of the readers who spend time at more partisan sites are wide-rangers, who are also more likely to end up at partisan sites on the opposite end of the ideological spectrum. Even more strikingly, segregation does not appear to be rising over time. Mr Gentzkow’s body of work reinforces the simple but reassuring point that what readers want most is to be informed.

_{Studies cited in this article:}

_{"Trading dollars for dollars: The price of attention online and offline", by Matthew Gentzkow, American Economic Review Papers and Proceedings, January 2014}

_{"Valuing new goods in a model with complementarity: Online newspapers", by Matthew Gentzkow, American Economic Review, June 2007}

_{"What drives media slant? Evidence from U.S. daily newspapers", by Matthew Gentzkow and Jesse Shapiro, Econometrica, January 2010}

_{"Media bias and reputation", by Matthew Gentzkow and Jesse Shapiro, Journal of Political Economy, April 2006}

^{Economist.com/blogs/freeexchange}