Template Tools
| Unproven claims about broadband |
| Wednesday, 03 March 2010 21:44 |
 40,000 people losing their jobs at Verizon know the job impact of broadband can be negative as well as positive. Verizon has the most advanced broadband network in the Western world and is massively cutting staff. The local bookstore that dies when people order from Amazon, the newspapers dying across the country as readers shift to the web, and eventually the schools displaced if on-line education actually grows are further evidence that broadband destroys jobs as well as creates them. Honest academics find a very mixed picture that suggests the net effect is very modest at best. The only honest answer looking at the data is Professor Shane Greenstein's "We don't know the effect of broadband on the economy." (I wish it were otherwise, of course.)Evidence-based medicine is transforming medical practice and research, undoubtedly savings lives, and I've been working to develop evidence-based policy. Washington is talking "data-driven decisionmaking but that's worthless if the data are corrupt and untested. There are claims being made in D.C. that range from unlikely to almost certainly untrue. Many originate from work paid for by the Bells then picked up and repeated by others who want to believe and won't examine the evidence. Huge economic and job impact of more broadband The data show otherwise, no matter how much I or advocates would like to believe this. Julius should never have said "that even modest increases in broadband adoption can yield hundreds of thousands of new jobs" based on a couple of Verizon paid studies essentially discredited by independent scholars. The author of one he cites, Bob Crandall, has several times has pointed out in prominent D.C. events that the 2003-2005 data in his work probably does not applytoday. During that period, broadband was first reaching many homes and businesses, many of whom couldn't connect at high speeds. In 2010, 95% of the U.S. can get landline and the remainder satellite. 65% of homes are connected and nearly any business with a major economic return is likely online already. It's a good thing to help the poor, elderly, and possibly even the illiterate get connected; there's no reason to believe these additional connections will have a major impact on the economy. The original Verizon funded claims have been strongly challenged by objective researchers. Shane Greenstein, the Elinor and H. Wendell Hobbs Professor at Northwestern told a recent NTIA hearing "We don't know the economic impact of broadband" and I know visited the FCC that day as well. Raul Katz of Columbia presented a paper in D.C. that pointed out broadband could either create or destroy jobs depending on which of several plausible assumptions one makes.  As Raul notes, "First, a "saturation" effect (i.e., when broadband adoption reaches high penetration levels nationally) might limit the economic impact of broadband. Second, ongoing research on the productivity impact of broadband indicates the potential for capital-labor substitution and consequently, the likelihood of job destruction resulting from broadband deployment. Third, since broadband tends to enable the outsourcing of jobs, a potential displacement of employment in the service sector from the area targeted for deployment might occur. Fourth, some job creation in the targeted areas could be the result of relocation of functions from other areas of the country, and therefore, should not be considered as creating incremental employment." http://www.elinoam.com/raulkatz/Dr_Raul_Katz_-_BB_Stimulus_Working_Paper.pdf Most dramatically, the most careful study of the economic impact of broadband was presented in D.C. in January. http://bit.ly/4SoNQL Jed Kolko finds a "positive empirical relationship" between availability of broadband in the U.S. early this century and economic growth. He goes on to point out that doesn't imply "broadband expansion causes economic growth." "The reverse might actually be true," he points out, "if broadband providers choose to offer or expand service in areas that are growing faster." During that period Bellsouth emphasized a "smart build" that looked at factors like the local economy. That might be a substantial confounding variable, although Kolko looked for effects like that and sees little evidence they are the explanation. "The overall relationship between broadband expansion and employment growth, as measured by the NETS, is positive." That's good news, and corresponds to my belief that broadband is a good thing. However, "both the average wage and the employment ratethe share of working-age adults that is employedare unaffected by broadband expansion. The economic benefits to residents appear to be limited. ... Broadband expansion is associated with no change in average pay per employee and a decrease in median household income. Broadband expansion has no statistically significant relationship with the employment rate. ... the economic development benefits of broadband are ambiguous." Jed, like I, believes broadband is a good thing and he was looking for a major economic impact. Unfortunately, the impact is not large enough to be clear from available data. Huge benefit from electronic health records and telehealth Legendary medical professor Jerome Groopman calls it "Obama's $80 Billion Exaggeration," and while that may be going too far the claims of $700B in savings rest on almost no evidence. Even if the returns from EHR are true, the broadband plan will do very little for their success because most medical facilities are already connected. (below). Actual speeds are typically 50% lower than advertised Bad Comscore data, cited by the FCC, has now been copied over the world and the refutation never seems to catch up. The 50% speed drop was higher than the results of the best survey in the world, Britain's OFCOM/SamKnows, although the U.S. has much less of a problem than the U.K. Cable is not affected by distance and is closer to promised speeds. The U.S. has twice as much cable as Britain and our advertisers are not as extreme. The same DSL line sold as "up to 8 megabits" in Britain is sold as "up to 6 megabits" in the U.S. so is more accurate. Our Verizon DSL line is rock solid at 3 megabits down and my Time Warner cable line is consistently close to the promised ten megabits. I’ve confirmed with the actual data on major networks that except for long loop DSL lines, speeds on any decent network tend to be within 20% of advertised. Even the feared "prime time dip" is generally exaggerated. Slowdowns are more often due to the servers, not the network. http://bit.ly/9gpAje bit.ly/cYu6Uv Any form of "demand stimulus" or broadband promotion works except bringing down the price Telcos learned years ago that promotion did little for broadband sales without a price break. This could be direct, in indirect in the form of a cheap computer, time payments, etc. So I looked closely at the claims that "promotion" made a difference in broadband take rates. There is no evidence that talking up broadband - demand stimulus - has ever worked. The most cited "study" - Connect Kentucky - falls apart when you look at the data and by their own criteria was worse than useless. (Availability went from 60-95% in the period. Adjusted for availability, the adoption in Kentucky was less than the national average.) I've asked expert after expert whether they have any solid evidence this kind of thing works. Nada. I'd like to think a little training would make a difference, but I can't find proof of that either. Most people in 2010 know what the Internet is, after all. Education: The planners cite data that kids with computers and broadband connections do much better in school as though the broadband connection is the primary reason the kids did well. It's far more likely that the more motivated (or less poor) kids are the ones with broadband, not that broadband is the primary cause of their success. Decades ago, Apple CEO John Scully told me that computers would transform education and solve most of the problems of the schools. Last I looked, U.S. schools aren't doing much better despite decades of computer use. Computers, with or without broadband, are good things but not transformative. There's a good discussion on how most interventions have little effect on schoolchildren http://nyti.ms/9SU5DF Inappropriate Claims The broadband planners are making several claims for benefits from the plan that will not be realized because of the plan. This is disingenuous. $513B saving from health records: Nearly all medical facilities are already connected at a speed fast enough for electronic health records. So whatever the benefits of health records, additional broadband won't have much effect. Attributing most of the $513B claimed benefit of electronic health records to anything in the broadband plan is unsupportable. Sufficiently fast lines are in place. Business web sites need more broadband like the plan would bring Julius credited web site orders for doubling the volume of a small business. "So it is that Blue Valley Meats in the small town of Diller, Nebraska doubled its employees and saw 40 percent growth by setting up a Web site and selling its beef online. But only once Diller got broadband." That's a good thing, but they could have done it previously by satellite without any problems. Nonsense. Most small business websites are hosted on servers like BlueHost, Rackspace, and GoDaddy, which cost as little as $10/month and are far more reliable than doing it yourself. There's no problem using a satellite connection - available almost everywhere - for updating the site and handling orders. Latency has little impact on creating a small business web site. Further examples welcome. |