Tumblelog by Soup.io
Newer posts are loading.
You are at the newest post.
Click here to check if anything new just came in.

May 30 2013

14:00

Which news moves stock prices?

Editor’s note: This summary of this interesting paper by Boudoukh et al. was written by Matt Nesvisky for the NBER (National Bureau of Economic Research) Digest. The paper asks the question: Through textual analysis and by tying news events to stock data, can we determine how news stories about companies are digested by the market?

In “Which News Moves Stock Prices? A Textual Analysis,” Jacob Boudoukh, Ronen Feldman, Shimon Kogan, and Matthew Richardson maintain that common business news sources, such as The Wall Street Journal and the Dow Jones News Service, contain many stories that are not relevant in terms of company fundamentals. They conclude that what is important for stock prices is the type and tone of the news. By applying advanced textual analysis to the actual language of news articles, they discern a strong relationship between information and stock price changes.

Boudoukh and his co-authors combine a dictionary-based sentiment measure, an analysis of phrase-level patterns, and a methodology for identifying relevant events for companies (broken down into 14 categories and 56 subcategories). Over the sample period of 2000-2009 for all S&P 500 companies, the Dow Jones Newswire produced over 1.9 million stories, but the researchers identify only about half of them as relevant events. This breakdown into “identified” and “unidentified” news makes a difference to the analysis, as does using a more sophisticated textual analysis, rather than a simple count of positive-versus-negative words.

Classifying articles into topics such as analyst recommendations, financial information, and acquisitions and mergers, the researchers compare days with no news, unidentified news, and identified news. They show that stock-level volatility is similar on no-news days and unidentified news days, which is consistent with the idea that the intensity and importance of information arrival is the same across these days. In contrast, the volatility of stock prices on identified news days is over twice that of other days.

Furthermore, the results are consistent with the idea that identified news days contain price-relevant information. Another finding is that deals and partnership announcements tend to have very positive effects, while legal announcements tend to have negative effects. Moreover, some topics, such as analyst recommendations and financials, are much more likely to appear on extreme return days. This suggests that different topics may have different price impacts.

Boudoukh, Feldman, Kogan, and Richardson conclude that their methodology may be useful for a deeper analysis of the relationship between stock prices and information, especially on the behavioral side. “There is a vast literature in the behavioral finance area,” they write, “arguing that economic agents, one by one, and even in the aggregate, cannot digest the full economic impact of news quickly. Given our database of identified events, it is possible to measure and investigate ‘complexity’ and its effect on the speed of information-processing by the market.

Here’s the abstract of the paper:

A basic tenet of financial economics is that asset prices change in response to unexpected fundamental information. Since Roll’s (1988) provocative presidential address that showed little relation between stock prices and news, however, the finance literature has had limited success reversing this finding. This paper revisits this topic in a novel way. Using advancements in the area of textual analysis, we are better able to identify relevant news, both by type and by tone. Once news is correctly identified in this manner, there is considerably more evidence of a strong relationship between stock price changes and information. For example, market model R-squareds are no longer the same on news versus no news days (i.e., Roll’s (1988) infamous result), but now are 16% versus 33%; variance ratios of returns on identified news versus no news days are 120% higher versus only 20% for unidentified news versus no news; and, conditional on extreme moves, stock price reversals occur on no news days, while identified news days show an opposite effect, namely a strong degree of continuation. A number of these results are strengthened further when the tone of the news is taken into account by measuring the positive/negative sentiment of the news story.

June 15 2011

17:35

October 06 2010

16:00

When people are willing to pay for “almost nothing”: The economic and emotional logic of web paywalls

Marco Arment is the developer behind Instapaper, the devilishly useful time-shifting tool for reading. Just as a DVR allows you to watch Mad Men when it’s convenient for you — say, at 6:45 p.m. Thursday instead of 10 p.m. Eastern on Sunday — Instapaper lets you read long-form prose when you’ve got the time and attention to devote to it. Come across a long interesting magazine piece in your daily web travels, but don’t have time to read it just then? Click Instapaper’s “Read Later” bookmarklet and it’ll be pulled into your iPhone (or Kindle or whatever) for offline reading when you’re on a plane or train. I love it.

But rather than just praise a terrific app, I want to point out a couple tweets of Marco’s that might tell us a little something about the paywalls we’ll see news organizations start erecting in greater quantities soon. Arment recently decided to start offering a paid model for Instapaper’s web service. He calls it an Instapaper Subscription, and it’s $3 for three months. What do you get for your $3? Arment is blunt:

Right now? Almost nothing, except knowing that you are supporting the Instapaper service’s operation and future feature development…

Some future features may be Subscriber-only, but please don’t buy a Subscription solely because you expect these exclusive features to be mind-blowing. They might be, depending on how easily your mind is blown, but I’d feel better if you bought the Subscription because you wanted to support Instapaper.

Now that’s a soft sell — pay me three bucks and I’ll give you roughly zero in return! But reaction to the move around the web has been overwhelmingly positive, as this tweet indicates:

I'm often surprised at how new things are received. Me: I now will accept money for almost nothing. Internet: Sounds great! Here you go!

He contrasts that with the often vociferous reaction some have to iPhone app developers who dare to charge a couple bucks for their work, rather than the expected price of $0:

Meanwhile.... iOS developers: Here is 6 months of work for $1.99. App Store reviewers: Lame, FAIL, should be $0.99, too expensive, useless!

An app is not a breaking story, and software is not journalism — but I think there’s some wisdom for news organizations here.

The economic value of your work is determined by the market, not wishes and hopes. Is it fair, in a cosmic sense, that people are willing to pay $3 for three months of “almost nothing” when it comes to Instapaper, but not pay $2 for an iPhone app that took an enormous amount of hard work? No! But prices aren’t set by principles of fairness: They’re set by the market. These are economic decisions, not emotional ones.

Is it fair that I’ll pay $20 a month for an email newsletter about book publishing, which is produced by a handful of people, but wouldn’t pay $20 a month for online access to The Boston Globe, which is the collective work of hundreds of talented journalists? No! But fairness doesn’t much enter into it. I’ve heard lots of journalists use words like “deserve” and “earned” when they describe why they want the paywalls to go up around their work. But the goal is to maximize the economic return on journalists’ work, not to act out of anger.

Requiring payment isn’t always fruitful than encouraging it. Imagine for a moment that instead of asking for subscriptions, Arment had put up an iron-clad, Times-UK-style paywall — pay $1 a month or else no more Instapaper for you. In that case, there’d probably be some users who’d pay up who wouldn’t with Arment’s soft sell.

But Instapaper has a classic competitive substitute good: a very similar service called Read It Later. Is Read It Later as good as Instapaper? I don’t think so — but it’s plenty good enough for the vast majority of people. A hard Instapaper paywall might raise revenue, but at the cost of driving away a huge chunk of customers off whom money might be made some other way — through advertising, say, or by being converted later to paying customers, or simply by promoting the app to their friends.

Most news is the very definition of a substitutable good. If CNN.com put up a paywall, MSNBC.com would be there waiting to collect the free traffic. Ditto The Washington Post and The New York Times. Most online news consumers aren’t looking for specific stories; they’re looking for something to occupy their time for a few minutes that makes them feel better informed. There will always be lots of free alternatives that can fill those duties. That’s why, while I’m happy to see news organizations experimenting with pay models, I’ve also been happy to see them setting reasonable expectations for the outcome and thinking a lot about how to maximize both revenue and audience.

Love and affection drives money. Why are people giving Marco Arment money for nothing? Because they love it. Check out these tweets:

In the online world of free, people need a damned good reason to fork over their money. It had better solve a problem, bring consistent delight, or otherwise earn devotion. A small but devoted audience can be worth more than a big, uncommitted one. How many people love their local newspaper?

September 20 2010

16:00

A warning to nonprofit news organizations: Government funding may not boost the bottom line much

At a time when some Americans are talking about increasing government support for journalism, here’s an interesting new study that adds a useful data point to the discussion: When governments provide financial support to nonprofit organizations, 73 percent of the extra money is counterbalanced by a decline in support from private donors. In other words, the value of government money received is decreased by a reduction in funds from elsewhere.

The paper is by Jim Andreoni of UC San Diego and A. Abigail Payne of Canada’s McMaster University, and it examines over 8,000 nonprofit organizations. The idea that government funding reduces private giving is not new, but this paper attempts to figure out why — and how — the trade-off occurs. Is it because private donors think that government grants eliminate their own need to give — the idea that they “already gave at the office” through their tax dollars? Or is it because getting government money causes nonprofits to relax, to reduce how aggressively they pursue outside money through fundraising?

Andreoni and Payne come down squarely on the side of the latter — it’s primarily nonprofits’ own reduction of their own fundraising efforts that lead to less outside support, not any change of heart by donors. When the government gives, nonprofits take that as an opportunity to cut back on fundraising, even though fundraising is highly cost-effective; the paper finds an average $5 return in gifts for every $1 spent on raising money. Reducing fundraising may save some cash in the short term, but it doesn’t appear to be a smart strategy.

If charity managers find fund-raising a “necessary evil,” or fear it may hurt their evaluation from charity watchdog groups, then a government grant will allow them to redirect efforts from fund-raising to providing charitable services. This means that after getting a grant, charities may simply cut back fund-raising.

The paper finds that for every $1,000 given through a government grant, nonprofits reduce their spending on fundraising by an average of $137. But that decrease leads to a drop of $772 in donor gifts. (The paper found that, contrary to the fears of some, government grants encourage outside donors to give instead of discouraging them — but the impact is small, only about $45 per $1,000 in government grants.)

In other words, adding it all together, $1,000 in government money only nets out to $410 in the end, on average.

The study didn’t look specifically at nonprofits engaged in journalism, and it’s difficult to apply its findings directly to the ongoing debate over government support for news. Check out the full paper for much more detail. But if I were in charge of a nonprofit news organization, here’s what I’d take away from Andreoni and Payne:

Government help is not a cure-all. Even setting aside the very legitimate arguments over the wisdom or ethics of government support for news, it doesn’t appear to be quite the financial boon some are foreseeing, at least for nonprofit organizations more broadly.

Fundraising is worth investing in. Andreoni and Payne say it’s surprising to economists that $1 spent on fundraising could lead to $5 in revenue, but it’s a robust finding that lines up with what the industry reports internally. They also point out that not every nonprofit approaches fundraising with the same sort of enthusiasm (the “necessarily evil”); those who find the task distasteful will pay for it in the pocketbook.

Success in one source of revenue can’t lead to the abandonment of others. The smartest nonprofit news organizations are busy trying to build a multi-pronged model for financial sustainability — often blending advertising, sponsorship, small individual donors, money from big foundations, content-sharing alliances, and more. Over-reliance on any one source is dangerous; just ask the publisher of a major metro newspaper about classified advertising circa 1995.

Photo by Thomas Hawk used under a Creative Commons license.

April 12 2010

21:40

ProPublica’s expensive story and deserved Pulitzer

Congratulations to ProPublica’s Sheri Fink, who just won the 2010 Pulitzer Prize for Investigative Reporting for her story about a New Orleans hospital in the immediate aftermath of Hurricane Katrina. (She shared it with Barbara Laker and Wendy Ruderman of the Philadelphia Daily News.)

We wrote about Fink’s terrific piece twice last fall. First, Zach Seward noted the huge cost of producing the story — $400,000 by one estimate — and the unusual cost-sharing between ProPublica, the Kaiser Foundation, The New York Times Magazine, and Fink herself. And I (gently!) tweaked the piece’s online presentation for not being as reader-friendly as it could have been.

February 04 2010

17:00

Is online news just ramen noodles? What media economics research can teach us about valuing paid content

The New York Times’ announcement that it would be charging for some access to its website, starting in 2011, rekindled yet another round of debate about paywalls for online news. Beyond the practical question (will it work?) or the theoretical one (what does this mean for the Times’ notion of the “public”?), there remains another question to be untangled here — perhaps one more relevant to the smaller papers who might be thinking of following the Times’ example:

What is the underlying economic value of online news, anyway?

Media economist Iris Chyi [see disclosure below] has a few ideas about this problem. An assistant professor in the School of Journalism at the University of Texas, she has been researching the paid-vs.-free, print-vs.-online conundrum since the late ’90s. Her research has consistently found that even while online news use continues growing, its preference lags behind that of traditional media. In other words: Even as audiences transition from TV/print news consumption to the web, they still like the traditional formats better for getting news, all other things being equal.

Now, this seemingly makes no sense: How could a format as clunky, messy and old-school as print “beat” such a faster, richer and more interactive medium on likability?

Chyi believes she found the answer in the economic principle of “inferior goods.” The idea is simple: When income increases, consumers buy more “normal goods” (think: steak) and fewer “inferior goods” (think: ramen noodles). When income goes down, the opposite occurs (again, all things being equal in economics terms). Inferiority, in this case, isn’t so much a statement of actual quality as it is of consumer perception and demand. If we get richer, our desires for steak go up and our desires for ramen go down.

What does this mean for journalism? “Users perceive online news in similar ways — online news fulfills certain needs but is not perceived as desirable as print newspapers,” Chyi said.

She and co-author Mengchieh Jacie Yang make this point through an analysis of data on news consumption gathered from a random sample of U.S. adults; their findings are published in the latest issue of Journalism & Mass Communication Quarterly, the flagship peer-reviewed journal for AEJMC. (See the related news release, overall highlights, and the full-text PDF). Chyi and Yang summarize their key findings as follows:

This analysis, based on data collected by the Pew Research Center in 2004, identified a negative relationship between income and online news consumption: When income increases, online news use decreases; when income decreases, online news use increases, other things (demographics, news interest, and/or other news media use) being equal — suggesting that online news is an inferior good among users. In contrast, the print newspaper is a normal good.

Such findings, at first glance, may surprise media scholars as well as online news professionals. After all, in communication research, no news products have been labeled as inferior goods before. In addition, major U.S. media companies have invested heavily in their online ventures, offering an array of interactive features and multimedia content — most of which are unattainable by print newspapers. It is therefore difficult to understand why online news could be an inferior good. Yet, from an economic perspective, “goods are what are thought of as goods.” Any product’s economic nature is determined by consumer perception and response. Based on this particular data set, which consists of survey responses collected from a national sample of online news users by a major polling institution in 2004, online news is an inferior good among users.

Clearly, the use of 2004 data is a limiting factor here (although the authors explain why more recent Pew surveys couldn’t be used for this kind of question). Yet, if we accept these findings, we’re left to unravel two mysteries: Why is online news perceived as an inferior good in the first place? And what should that mean for the future of web journalism?

On the first question, there are at least several possibilities, as Chyi suggests. Maybe the computer screen just isn’t an enjoyable reading device. (And how might that compare with smartphones and e-readers?) Or maybe online newspapers still have content/design problems — think of all the ads for teeth whitening and tummy tightening, not to mention the general lack of contextual cues afforded by print. Or maybe it’s simply because online news is free — and, as behavioral economics research has indicated, sometimes consumers perceive higher-price products as more enjoyable. In any case, as Chyi puts its: “More research, as opposed to guesswork or wishful thinking, on the perception of news products is essential.”

Then there’s the second question: What does this suggest about the future of online news? Perhaps nothing too dire, as people still do pay for ramen noodles when it suits them — when the price, convenience, or alternatives make ramen noodles the preferred choice. This isn’t to suggest that consumers invariably will pay for online news, but rather that they might if the perception calculation is right.

The key here is to recognize that consumers are rapidly adopting online news not necessarily because they prefer the medium to print, but because online news is “good enough” — cheap, convenient, flexible, and sufficient to satiate our information cravings. (This takes us into territory related to disruptive innovations and fidelity vs. convenience — interesting stuff, but something for a later post.) But the danger is in taking a “platform-neutral” approach if that leads one to assume that content value remains constant between print and online — that, basically, you can charge for content either way. Chyi suggests that is like trying to market ramen noodles as steak: Newspapers do so at their peril.

So, what does all of this say about the Times and its paywall? Perhaps not much because, after all, “the Times is the Times.” Yet, the notion of online news as an inferior good highlights a few salient points for thought: (1) news usage doesn’t always correlate with preference, counterintuitive as that is; (2) publishers hoping to charge for niche content need to understand where their offering fits in the normal-inferior goods relationship, and how that should affect pricing and marketing strategies; and (3) there’s a critical need for R&D to help us grasp why consumers perceive online news as inferior, and how that perception might vary among different demographics of users and/or according to different types of news content.

In the meantime, enjoy your ramen noodles.

[Disclosure: Chyi and I have collaborated on several research projects through her Media Economics Research Group in the School of Journalism at the University of Texas — including a recent peer-reviewed article on newspapers' effectiveness in penetrating the local online market (PDF). Also, she's currently a member of my dissertation committee.]

Photo of ramen by Broderick used under a Creative Commons license.

January 23 2010

18:19

January 22 2010

07:50
Older posts are this way If this message doesn't go away, click anywhere on the page to continue loading posts.
Could not load more posts
Maybe Soup is currently being updated? I'll try again automatically in a few seconds...
Just a second, loading more posts...
You've reached the end.

Don't be the product, buy the product!

Schweinderl