Browse: Home / Most US investors get PR wire releases 80 seconds after Wall Street
Learn about IR Web Report's Online IR Audits

Most US investors get PR wire releases 80 seconds after Wall Street

By Dominic Jones on August 3, 2010

  • Email

A SIMPLE analysis of 100 company news releases shows that the average delay between when PR wire services release the information to Wall Street and when most non-professional investors actually have access to it on Yahoo! Finance is just over 83 seconds.

A difference in access of that magnitude is unacceptable in today’s market where company news is routinely the subject of heavy trading in the pre- and after-hours markets. It’s especially galling because companies and regulators have free or low-cost ways to easily close the access gap between Wall Street and mainstream investors.

Growing public awareness of the uneven nature of company news dissemination is likely to undermine investor confidence in the fairness of the markets and  diminish their willingness to invest. No one wants to participate in a market where some get earlier access to information than others.

Real-time web increases public awareness

That there are latencies in the PR wire dissemination system is nothing new. It is inherent in the Internet protocols and computer systems that most PR wires rely on to distribute and publish company news on a myriad different websites. Many factors can affect the time it takes for information to travel over the web, including network congestion, the size of the releases, and the speed at which the receiving systems process and publish the information.

While PR wire services like Business Wire boast about their patented simultaneous distribution technologies, the fact is they cannot guarantee that an investor waiting for a company earnings release on a public investment website will be able to access it at the same time as a professional trader looking at a Bloomberg terminal. It depends on the factors listed above, and in Business Wire’s case, on the recipient system having special software installed to decipher pre-delivered encrypted releases.

However, these inherent latencies have become much more noticeable and relevant to the investing public in the two years since websites like Google Finance and others started publishing real-time stock prices. Prior to that, prices were typically delayed by 15 minutes, so investors on public websites were less aware of Wall Street traders reacting to news that they themselves had not yet seen.

Awareness of uneven news dissemination has also grown with the advent of real-time information sharing services like StockTwits, where traders with access to professional terminals routinely post company news before the PR wire releases make it to popular investment websites.

Quantifying the latency of PR wire distribution

Trying to quantify the delay between when PR wires release company news and when it actually is published on public finance websites has been difficult and imprecise. This is because investment websites show the PR wire service time stamp on the releases they publish, even though the time shown is earlier than when the news actually appears on the public website.

Yahoo! Finance logoHowever, I recently discovered that Yahoo! Finance offers an earnings news feed that shows both the PR wire services’ time stamps and when the news is actually published in the feed. Yahoo! Finance is by far the biggest investor site on the web with an audience of 26.25 million unique visitors, by one recent estimate. Its users include most retail investors and also many financial professionals who use it alongside their professional services or from home.

By comparing the PR wire services’ time stamps to Yahoo! Finance’s own time stamps one is able to easily calculate the latency (see screenshot below). I did this for 100 news releases over a period of several days and at random times to get a fairer picture of the delays.

I chose random times and days to avoid potentially skewing the results, assuming that it takes longer for Yahoo! to process releases during busy periods, such as between 4:00pm and 5:00pm ET when many companies are issuing releases.

I also checked both full earnings releases, which are typically more than 1,000 words, and shorter advisories about upcoming earnings announcements. I did this because large files probably take longer to be processed than shorter ones.

Finally, I didn’t focus on any particular PR wire service, but took whatever was moving over the PR wires at the time. In the end, the results included releases from PR Newswire, Business Wire, Globe Newswire, Marketwire and CNW.

As I’ve already said, the average delay was 83 seconds. The fastest was 24 seconds and the slowest was 237 seconds or almost 4 minutes. The median was 80 seconds.

I didn’t calculate which PR wire was fastest or which was slowest because that wouldn’t be fair given that some PR wires were underrepresented in the list.

However, latencies were all over the map for all of the PR wires. For example, the speeds for Business Wire releases ranged from 24 seconds (the best) to 130 seconds, which was the 10th slowest. If anyone wants the results, I’ve posted a summary in a published Google Spreadsheet.

You can do your own research using the source code of the same feed I used. You have to look at the feed source, which you access in different ways depending on your browser, but right clicking on the screen should give you an option to see the source code.

These findings have important implications for companies and regulators. I’ll explore some of the issues in an upcoming post. In the meantime, let me know what your experiences are.


Dominic Jones

Dominic (bio & disclosures) is IR Web Report‘s founder and an online investor relations consultant. He advises leading public companies and investor relations service providers worldwide on using the web for disclosure, engagement and profile building. You can contact him via the contacts page.

Posted in , , | Tagged , , , , , | 11 Responses

  • itserich

    This is important. Another issue is the amazing delay in court rulings. Lawyers will whip out their blackberries and the big investors find out in minutes while the average person may wait hours or days for hard facts. Even an investor trying to stay informed has little recourse, Pacer alerts are generally not available. Courts should issue rulings when the markets are closed.

  • itserich

    This is important. Another issue is the amazing delay in court rulings. Lawyers will whip out their blackberries and the big investors find out in minutes while the average person may wait hours or days for hard facts. Even an investor trying to stay informed has little recourse, Pacer alerts are generally not available. Courts should issue rulings when the markets are closed.

  • Pingback: Time stamping by stock exchanges « Prof. Jayanth R. Varma’s Financial Markets Blog

  • http://twitter.com/dougchia Douglas K. Chia

    Amazing to think that back when Reg. FD was adopted in the late 1990s, we were measuring delays in business days whereas now we're measuring it in seconds!

  • newrulesofinvesting

    this is a great post. I like your analysis.

  • Sacandaga

    Dominic:
    The “latency” you describe is, as you say, nothing new, but it is likely (absent evidence to the contrary) a factor of something a lot less sinister than deliberate acts by the newswire services in favor of Wall Street.
    Anyone familiar with basics of data-network transmission knows that the effective transmission speed for any given file is a sum-of-the-parts exercise. Data networks break down files into lots (hundreds or thousands, depending on file size) of little packets, transmit each packet separately by what may be hundreds of different routes, reassemble the file when the slowest packet reaches its destination, place the reassembled file back together and error-check it, and then signal its arrival to the recipient’s server.
    That’s why transmission speed can vary significantly even for two copies of the same file sent from the same place, and likely a part of why the “latency” you measured varies to the degree you detected.
    But unless you have specific evidence of continued actions by newswire services (i.e., separately timed file transmissions or access to alternate servers) favoring Wall Street, it’s likely other factors allow the “have-more” crowd to get the news sooner than “have-less” folks — that would be not just the processing horsepower and software on a Bloomberg terminal, but more likely the size and speed of the data “pipes” connecting traders’ terminals to their data-network providers.
    Since Wall Street has invested in the fastest-possible network access, internal networks, AND terminals to make sure its people get those server signals as quickly as possible, it stands to reason that their traders will get the news faster than, say, someone at home or in a small office relying on a standard-grade DSL or cable-modem connection, perhaps further handicapped by a wireless vs. wired router connection, leading to, say, a standard HP laptop. Whether those technology differences account for access differences of a minute or more is unknown absent more in-depth investigation, but they are likely the principal factor.
    All of which is to say that it would be ridiculous to for a newswire service to guarantee (or to be held responsible for) equal end-user access to the information it distributes, because it does not control transmission once the content leaves its system. The only thing they can and should be held responsible for is making one and only one distribution of a release, and only via the public data network.
    That is not to say that there are no public-policy issues here. But the concern should perhaps be raised first with the FCC rather than the SEC. That agency is under an all-out lobbying attack by data network providers to in essence permit the creation of multiple parallel data networks, priced by speed and capacity, which could lead to even greater wealth-related differences in access to time-sensitive financial information. The providers say they would not allow market-sensitive public information to be accessible based on price, but it would be far easier for clever technicians to exploit public-network speed differences for that purpose if such differences existed.

  • Pingback: Netflix latest to join web disclosure trend | IR Web Report

  • Pingback: IR websites resemble “wire services” as Q4 taps APIs | IR Web Report

  • Pingback: How Q4 Integrates Twitter and Facebook APIs with Corporate Websites | Q4 Blog

  • Pingback: How Q4 Integrates Twitter and Facebook APIs with Corporate Websites | Q4 BLOG BETA SITE

  • Pingback: Automatisierte Investor Relations Meldungen erreichen Social Media | Online Investor Relations