Donate
Independent, objective, nonpartisan research

R 110JKR

Authors

R 110JKR

Tagged with:

Publication PDFs

Database

This is the content currently stored in the post and postmeta tables.

View live version

object(Timber\Post)#3711 (44) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(5) { ["_wp_attached_file"]=> string(12) "R_110JKR.pdf" ["wpmf_size"]=> string(6) "519169" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(107798) "Does Broadband Boost Local Economic Development? January 2010 Jed Kolko with research support from Davin Reed Supported with funding from The David A. Coulter Family Foundation http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 2 Summary “Broadband” refers to high -speed Internet service that —unlike dial-up modem service— is always “on.” This technology has become widely available throughout the United States. It is essential for businesses, and well over half of American households have broadband access at home. In recognition of its importance, public investment in broadband is surging. The American Recovery and Reinvestment Act (ARRA) of 2009 allocated $7.2 billion for broadband investment and commissioned a National Broadband Plan to promote universal access, foster economic development, and achieve additional potential benefits through thi s technology. Several California programs also support the expansion of broadband access, especially in areas where availability is lagging. This report assesses whether policies to raise broadband availability will contribute, as hoped, to local economic development. Our analysis relies on the fact that broadband technology has diffused unevenly throughout the United States, thus allowing us to compare economic indicators between areas with greater and less growth in broadband availability. Using broadban d data from the Federal Communications Commission and economic data from several government and proprietary sources, we examine broadband availability and economic activity throughout the nation between 1999 and 2006. Our analysis indicates a positive rel ationship between broadband expansion and economic growth. This relationship is stronger in industries that rely more on information technology and in areas with lower population densities. Although the evidence leans in the direction of a causal relations hip, the data and methods do not definitively indicate that broadband caused this economic growth. T he economic benefits to residents appear to be limited . Our analysis indicates that broadband expansion is also associated with population growth and that both the average wage and the employment rate —the share of working -age adults that is employed —are unaffected by broadband expansion. The economic benefits to households are thus more ambiguous than they would be if employment growth also led to an increas e in wages or the employment rate. We also found that expanding broadband availability does not change the prevalence of telecommuting or other home -based work. Of course, local employment growth might still raise property values and the local tax base , but in the absence of more direct benefits for residents in the form of higher wages or improved access to jobs, we can only say that the local economic development benefits of broadband are mixed. Broadband expansion may of course offer other social or economic benefits, such as improved health care delivery. Although our study does not examine such effects, we briefly review the limited evidence of other benefits in the conclusion of the report. Contents Summary 2 Tables 4 Figures 5 Introduction 6 Broadband Policies and Goals 8 The Federal Economic Stimulus and the National Broadband Plan 8 Broadband Policy in California 9 Broadband Availability and the Digital Divide 11 Unserved and Underserved Areas 11 Limitations of Available Broadband Data 14 What We Know about Broadband Availability 14 Why There Is a Digital Divide in Broadband Availability 15 Research Questions and Approach 17 Questions Addressed in This Report 17 Research Strategy 19 Questions This Report Does Not Address 21 Research Findings 22 Broadband and Local Overall Employment Growth 22 Broadband and Employment Growth across Industries and Places 23 Does Broadband Expansion Cause Employment Growth? 24 Broadband and Household Outcomes 25 Implications and Conclusion 28 What the Results Mean for Broadband as Local Economic Development Policy 28 Other Possible Benefits of Broadband 30 Broadband: The Next Generation 31 References 32 About the Author 34 Acknowledgements 34 All technical appendices to this paper are available on the PPIC website: http://www.ppic.org/content/pubs/other/110JKR_appendix.pdf http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 4 Tables Table 1. Broadband speed illustration 12 Table 2. Number of broadband providers by ZCTA 15 Table 3. Broadband and economic outcomes, 1999 –2006 22 Table 4. Broadband and industry employment growth, 1999 –2006 23 Table 5. Summary of findings 28 http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 5 Figures Figure 1. Broadband Availability in California, August 2009 13 http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 6 Introduction Since the mid -1990s, when businesses and households began to use the Internet, observers have been trying to predict —and more recently to assess —the effects of this relatively new mode of communication. Some of the early predictions about the Internet and related technologies, such as causing the “death of cities,” have clearly not materialize d. 1 The Internet has transformed many areas of life, providing individuals and businesses with easy and instant access to communication, information, and entertainment. But it may take years before the Internet —like many other new technologies —exerts a quantifiable effect on business productivity. It takes considerable time to develop applications and adjust business processes and organizations to take full advantage of new tec hnologies. But we still know very little about the economic effects of these technologies. 2 Nonetheless, governments need to assess whether new technologies are likely to offer benefits in the public interest and whether the private sector will make the technology available and affordable enough to support public policy goals. Occasionally, governments will deem a new technology essential for achieving these goals and will support universal access, even in areas that are expensive to serve. The federal government and the state of California, as well as other state and local governments , have made universal access to broadband service a public policy goal —the specific objective is to close the digital divide in broadband availability. 3 The phrase “digital divide” can refer to geographic inequalities in availability or to gaps in broadban d adoption owing to income, race/ethnicity, education, or other inequalities in access or skills that affect the ability of individuals or businesses to take advantage of broadband’s capabilities. 4 The federal and state broadband initiatives presume multiple economic and social benefits will accrue from increasing broadband access. Local economic development ranks high among these benefits . Other benefits include improvements in health care delivery, access to education, energy efficiency, civic participation, and public safety. To date, the evidence on the extent to which broadband provides any of these benefits has been quite limited. To help promote the goal of universal access, the federal government allocated $7.2 billion for broadband investment and also commissioned a National Broadband Plan as part of its 2009 American Recovery and Reinvestment Act (ARRA). These federal “stimulus” funds provided a large addition to existing state and lo cal efforts to support broadband access. In this report, we are particularly interested in the relationship between broadband and economic development. Previous research has not assessed whether broadband expansion causes economic development, nor has it examined who benefits from increased economic activity; there ha s been little research on the effect of broadband on outcomes other than employment, output, and income. Our analysis relies on the fact that broadband technology has diffused unevenly throughout the United States, thus enabling us to compare economic indi cators between areas with greater and less growth in broadband availability. In this report, we examine broadband availability throughout the nation between 1999 and 2006. Our parameters are partly determined by our 1 George Gilder, Forbes ASAP , February 27, 1995. Quoted in Mitchell Moss, “Technology and Cities,” Cityscape 3:3 . 2 “The Broadband Myth,” The Economist , May 23, 2008. 3 While broadband can include satellite and high -speed mobile data services, the vast majority of households and businesses that subscribe to broadband receive their service over cable, digital subscriber line (DSL), or other technologies that use physical wires or c ables to connect end- users to the Internet. Discussions about broadband availability usually re fer to wireline technologies. 4 “Digital divide” can refer to inequalities in the availability or adoption of other technologies as well. See Gunkel (2003). http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 7 data: Federal Communications Commission (FCC) data on broadband are available beginning in 1999, and our source for detailed employment data ends in 2006. We answer four questions: 1. Does employment grow faster in areas with greater broadband expansion? 2. Does the relationship between broadband and employment differ by industry or across places? For example, is it stronger for industries that are more reliant on technology or that use workers who are more technically knowledgeable? Is it stronger in places that are more isolated or in those with more amenities? 3. If there is a positive relationship between broadband and employment growth? Does broadband expansion cause employment growth? 4. If broadband does boost employment, who benefits? Is employment growth accompanied by a greater likelihood of employm ent, higher pay, increased income, or greater flexibility to be able to work from home? Although other studies have examined broadband and economic growth, this report offers more definitive answers by using richer data and more -refined methods. We also ex amine questions left unanswered in previous work, such as whether broadband actually causes economic growth, who benefits from this growth, and whether broadband increases telecommuting and other forms of home -based work. Of course, economic development is only one of many policy concerns that must be considered when targeting broadband investments, and in our concluding chapter we briefly review the limited evidence available on other potential benefits of this technology, including improved health care de livery and overall consumer welfare. We begin our discussion in the next chapter with a description of the national and California policies designed to increase broadband availability. We then describe the extent of broadband availability, explain why some places have better broadband access than others, and discuss the limitations of the available data. We then present our research questions, the methods we use to answer them, and the main findings of our study. We conclude with a discussion of the policy implications of our findings and then briefly review other possible effects of broadband that are outside the scope of our analysis. A technical appendix offers further details about related research and our data, methods, and results. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 8 Broadband Policies and Goals Broadband policy at both the national level and in California is designed to achieve multiple goals, most prominently economic development. The federal government, throug h Congress, the executive branch, and the Supreme Court, has shaped broadband policy through impo rtant regulatory decisions. 5 And today , federal public investment in broadband is surging. ARRA, the $787 billion economic stimulus package passed in 2009 , allocated $7.2 billion to broadband investments, with a particular focus on increasing broadband availability in targeted areas to improve economic development. 6 Well before the current federal investment in broadband, state and local governments have subsidized and in some cases directly provided broadband services. California has several programs in place that subsidize the demand and supply of broadband. The stated goals of these policies —described in detail below —include job creation and economic growth, civic participation, public saf ety, education, health care, and access to government services. Like federal policies, some California initiatives are more narrowly targeted toward underserved areas and economic development goals, while other programs are focused on a wider set of concer ns, such as health care and education. ARRA also requires the development of a National Broadband Plan to guide broadband policy beyond the federal stimulus funds. Th is p lan is supposed to be wide -ranging, covering numerous technologies, policy options, and goals. In sum, broadband policy at the federal level and in California aims to achieve a wide range of goals, with a particular emphasis —especially in the federal broadband stimulus —on raising broadband availability in unse rved and underserved communities in an effort to create jobs and stimulate economic growth. The Federal Economic Stimulus and the National Broadband Plan Signed into law in February 2009, ARRA directs two federal agencies, the Department of Agriculture’s Rural Utilities Service (RUS) and the Department of Commerce’s National Telecommunications and Infrastructure Administration (NTIA), to grant or lend $7.2 billion for broadband deployment and other broadband projects. 7 ARRA directs both agencies to conside r the effect on economic development when awarding funds. RUS will offer grants, loans, and loan guarantees worth $2.5 billion for broadband infrastructure projects in rural areas that lack “ sufficient access to high speed broadband service to facilitate rural economic development.” 8 5 The 1996 Telecommunications Act, the 2005 Supreme Court decision in National Cable & Telecommuni cations Association et al. v. Brand X Internet Services et al., and related FCC rulings helped create the national regulatory framework in effect today. Under this framework, telephone, wireless, television, and Internet providers are able to compete in an y market against each other, but broadband providers are not required to give competitors wholesale access to their infrastructures that would enable competitors to resell services to co nsumers. NTIA will award grants worth $4.7 billion to states, non- profits, and broadband providers for a wider set of broadband projects —not just infrastructure projects and not just in rural areas. While economic growth and job creati on are explicit goals of NTIA’s funds, as is raising broadband access in unserved and underserved areas, so is improved broadband access for schools, medical facilities, other community institutions, and 6 Of the $4 billion to be allocated in the first round of fun ding, 75 percent has been targeted toward infrastructure projects. 7 Applications for broadband stimulus funds are being accepted and reviewed in multiple rounds. First round applications were due in August 2009; up to $2.4 billion of the RUS funds and $1. 6 billion of NTIA funds will be awarded in this round. Subsequent rounds will award the remainder of the $7.2 billion by September 2010. Stimulus funds will be allocated by RUS through the Broadband Initiatives Program (BIP) and by NTIA through the Broadba nd Technology Opportunities Program (BTOP) . 8 ARRA, (H.R. 1) Title I, p. 4. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 9 public safety agencies. Together, these two components illustrate the prominence of economic development and job creation as goals for the broadband stimulus funds, among other goals. ARRA also requires the FCC, the national telecommunications regulator, to present a National Broadband Plan to Congress in F ebruary 2010 to guide future broadband policy. The P lan is supposed to “seek to ensure that all people of the United States have access to broadband ca pability” and is an acknowledgement that the broadband stimulus funds alone are “insufficient to support national broadband deployment.” 9 The Plan should recommend goals and benchmarks for a national broadband strategy and for achieving eleven enumerated policy goals, including job creation and economic growth, entrepreneurial activity, and other specific tar gets such as consumer welfare, civic participation, public safety, education, and health care delivery. 10 Broadband Policy in California The plan will thus cover a wider range of interests and technologies than the broadband stimulus funds, which are geared toward infrastructure, technol ogy mapping, and related activities. Well before the passage of ARRA, s tates and localities—including the state of California and some of its cities and counties —have made broadband investments through subsidies and direct p rovision. 11 As with federal broadband initiatives, California’s broadband policy is focused on numerous goals, with economic development prominent among them. 12 For many years, the state’s primary broadband subsidy program has been the Teleconnect Fund, esta blished in 1996, which pays half the cost of Internet access for qualified schools, libraries, community organizations, and other nonprofits, funded through a statewide fee on telephone service. The program has grown from $33 million in FY 2008– 2009 to $47 million (proposed) in FY 2009 –2010 and $67 million (proposed) in FY 2010– 2011. 13 Two newer programs have expanded the state’s involvement in broadband infrastructure deployment. In 2006, the California Public Utilities Commission (CPUC) created the Califo rnia Emerging Technology Fund (CETF), an independent nonprofit foundation. CETF has received funding for its first five years through a $60 million contribution from AT&T and Verizon, a condition of their respective mergers with telecommunications compani es SBC and MCI. CETF’s mission is to achieve “ubiquitous access to broadband and advanced services in California, particularly in underserved communities.” The foundation seeks to promote broadband availability in rural areas, affordability and adoption in urban poor and other disadvantaged areas, and accessible technology for people with disabilities. CETF hopes to disburse $240 million over five years, supplementing its own seed money with matching funds. As of December 2008, it had awarded $20 million i n grants. The largest award helped support the California Telehealth Network (CTN), which electronically connects hundreds of clinics and hospitals —primarily in rural areas and tribal lands —to medical centers. CETF’s support also helped CTN win a larger grant from the FCC’s Rural Health Care Pilot Program. 9 ARRA (H.R. 1), Section 6001 (k)(2), p. 402; FCC Notice of Inquiry 09- 31, 2009, paragraph 6. 10 The FCC’s Notice of Inquiry reviews these and other current FCC activities that relate to broadband. 11 Although the major current initiatives are sponsored at the federal and state levels, some localities in recent years have provided broadband directly t hrough public municipal Wi- Fi (wireless) networks, fiber-optic networks, or other broadband technologies. Kolko (2007) discusses local initiatives in California. 12 In addition to the subsidy and investment programs described in this section, California has set broadband policies through the final report of the California Broadband Task Force, created by Executive Order S -23- 06 in 2006, and the 2006 Digital Video and Infrastructure Competition Act (AB 2987), which created statewide video franchises and estab lished the CPUC’s responsibility for collecting and mapping broadband data. 13 Actual and proposed budgets are available at the CTF website, www.cpuc.ca.gov/PUC/Telco/Public+Programs/CTF/CTFList.htm . http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 10 The second new state program is the California Advanced Services Fund (CASF), which is providing $100 million over two years to subsidize broadband deployment in unserved and underserved areas; successful applicants are awarded 40 percent of total project costs for broadband infrastructure in these areas. CASF was authorized by the CPUC in 2007 and implemented in 2008 under SB 1193 (Padilla); the first projects were funded in November 2008. Whereas CETF ha s a broad strategy that includes infrastructure availability, adoption, applications, and technology literacy, CASF is more narrowly focused on promoting infrastructure availability in the largely rural parts of the state that lack adequate broadband. Thus , the main goal of CASF is quite similar to the objectives of the portion of the broadband stimulus funds targeted toward broadband availability in rural areas. The funds currently available through state programs —the Teleconnect Fund, CETF, and CASF—are small relative to the sums California might expect to receive under the federal broadband stimulus program. While the state programs have budgets or plans to disburse tens of millions of dollars annually, California’s share of the federal broadband funds w ould be over $900 million. 14 Thus, an essential element in California’s current broadband strategy is to win federal broadband stimulus funds. Numerous cities, counties, nonprofits, and companies in California were among the first -round applicants for federal funds: Their grant and l oan requests totaled $1.4 billion. 15 CPUC was a winner in the first round of the separate broadband mapping funds competition, receiving a $2.3 million grant in October 2009 . 16 California’s expectations, like those of the federal government, are that broadba nd will contribute to economic development. The final report of the California Broadband Task Force, for example, lists “economic and community development” first among the ways that broadband is likely to affect California, citing the potential effect of broadband on job creation in high- tech and other industries as well as on employment in rural areas. 17 The state also expects broadband to lead to increased telecommuting and environmental benefits. I n establishing its definition for “underserved” areas for CASF grant eligibility, the CPUC chose the minimum speed that would support telecommuting. 18 In turn, t he California Air Resources Board cites telecommuting as one way the state government can encourage its own employees to “decrease their individual carbo n impact.” 19 14 CETF’s “Summary of ARRA Proposals” estimates California’s “fair share” to be 13 percent of the federal total, proportional to the state’s share of the national population. 15 These applications have been submitted for the first round of $4 billion in federal funds, of which California’s 13 percent “fair share” would be $520 million —roughly one -third of the amount that California -based organizations have applied for. Applicants are listed at www.ntia.doc.gov/broadbandgrants/applications/search.cfm (viewed on October 12, 2009). The total request in California -based applications was reported in CETF’s “Summary of ARRA Proposals.” 16 The NTIA grant comes through the State Broadband Data and Development Grant Program, the program authorized by the 2008 Broadband Data Improvement Act and funded with up to $350 million from ARRA. 17 California Broadband Task Force, “The State of Connectivity,” 2008. Viewed on October 14, 2008, at www.calink.ca.gov/pdf/CBTF_FINAL_Report.pdf . Governor Schwarzenegger’s official response to the request for comments on the broadband stimulus also notes the importance of economic development: “ … with out any broadband service, [an area’s] residents are excluded from the economic development and social- economic benefits of broadband, including tele -education, telemedicine, access to health care information, access to government services and benefits, an d more.” See California response to BIP/BTOP request for comment, www.recovery.ca.gov/Content/Documents/California_Response_to_NTIA_on_ BB_Stimulus_Grants_4-13 -09.pdf , p. 12. 18CASF chose minimum speeds of 3 megabits per second downstream and 1 megabit per second upstream, arguing that at these speeds telecommuting becomes feasible, and lacking rapid enough connections to work from home qua lified an area as “underserved.” See Table 1 and California’s response to BIP/BTOP request for comment, www.recovery.ca.gov/Content/Doc uments/California_Response_to_NTIA_on_BB_Stimulus_Grants_4 -13 -09.pdf . 19 California Air Resources Board, Climate Change Draft Scoping Plan , June 2008, page 12. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 11 Broadband Availability and the Digital Divide California and federal policies seek to close the digital divide in broadband availability, placing a particular emphasis on unserved and underserved areas. ARRA and CASF define unserved and underserved in terms of whether broadband service is available and, if so, at what speed. CETF also mentions “ubiquitous access” and supporting “underserved communities” in its mission statement. Targeting “unserved” and “underserved” areas requires defining th ese terms and identifying which areas qualify. Recent improvements in data collection, and planned improvements funded by ARRA, will provide policymakers with much better information on broadband availability. However, for analyzing past trends in broadban d availability and the relationship between broadband and local economic development, we must rely on the less detailed data collected historically by the FCC. Unserved and Underserved Areas ”Unserved” and “underserved,” as defined for the federal stimulus funds, are an important policy statement about what constitutes adequate broadband availability. “Unserved” areas lack broadband service offering downstream speeds of at least 768 kilobits per second (kbps) and upstream speeds of 200 kbps. 20 These minimum speeds are lower than typical digital subscriber line (DSL) and cable broadband service today, which often advertise top downstream speeds of 1.5 to 6 megabits per second (mbps) and top upstream speeds of at least 768 kbps ( Table 1). “Underserved” areas ha ve either only partial service coverage at those speeds, no service at a higher speed threshold of 3 mbps downstream and 1 mbps upstream, or low adoption rates. 21 To assess the extent of the digital divide and to help identify unserved and underserved area s, the FCC recently improved its data collection efforts, and ARRA is allocating funds for additional mapping and surveying efforts. Since December 2008, broadband providers have been required to report subscribership levels by Census tract at different sp eeds of service through the FCC’s “Form 477” data collection process. (Form 477 data is about subscribership, and availability is inferred, based on the location of subscribers. This could result in an undercount of availability in areas where a provider has infrastructure allowing it to provide broadband service, but no subscribers.) As noted above, ARRA also seeks to provide information on unserved and underserved areas. It provides funding for the development of maps showing broadband availability and calls for the development of “a comprehensive nationwide inventory map of existing broadband service capability and availability” by February 2011. Finally, the 2008 Broadband Data Improvement Act requires the FCC to conduct consumer surveys on pricing, sp eed, adoption, and online behaviors, and to collect international data on broadband speeds and prices. 20 “Downstream” refers to data sent “down” from the Internet to an end -user, such as a downloaded file or a received email. “Upstream” refers to data sent “up” to the Internet from an end -user, such as an uploaded file or a sent email. 21 Unserved areas consist of contiguous Census blocks where at least 90 percent of households lack access to terrestria l (i.e., not satellite or mobile phone) broadband service offering speeds of at least 768 kbps downstream and 200 kbps upstream. Underserved areas consist of contiguous Census blocks where either (1) 50 percent of households or fewer have access to terrest rial broadband service of at least 768 kbps downstream and 200 kbps upstream; (2) no broadband service offers speeds of at least 3 mbps downstream and 1 mbps upstream; or (3) 40 pe rcent of households or fewer subscribe to broadband. Note that the standard of 3 mbps downstream / 1 mbps upstream is the level that CASF believes is necessary to support telecommuting, as mentioned in the previous section. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 12 TABLE 1 Broadband speed illustration Downstream speed Relevance Approximate time to download 3 megabyte pop song 56 kilobits/second (kbps) Top speed of dial -up modem service 7 minutes 200 kbps Minimum speed to qualify as “high-speed” (broadband) under historical FCC definition 2 minutes 768 kbps Minimum speed for an area not to be “unserved” according to ARRA 30 seconds 1.5 megabit s/second (mbps) Lo wer speed tier advertised for many broadband (DSL or cable) services 15 seconds 3 mbps Minimum speed for an area not to be “underserved” according to ARRA, and minimum speed CASF considers necessary for telecommuting 8 seconds 6 mbps Upper speed tier advertised for many broadband (DSL or cable) services 4 seconds 10 mbps Speed of advanced services available in some areas, including much of metropolitan southern California 2 seconds 100 mbps Speed of advanced services available in very limited areas, including parts of metropolitan Sacramento ¼ second NOTE: 1 byte equals 8 bits. Some states, including California, have already undertaken their own mapping initiatives. Based on infrastructure data collected from providers, Californi a publishes maps of broadband availability throughout the state, disaggregated into several speed tiers. 22 Maps dated August 10, 2009, show large rural and mountainous areas in the state without broadband access. The fastest service —more than 100 mbps— is a vailable in parts of the Sacramento metropolitan area; service at speeds of 10 –100 mbps is available throughout much of urban southern California, as well as Bakersfield and Napa and Solano Counties; and speeds of 5 –10 mbps are offered in most of the Bay A rea, including Silicon Valley (Figure 1). These differences demonstrate that even among places with broadband availability, speeds can vary considerably. And much of the state appears to have no service at speeds of at least 500 kbps, thus meeting the defi nition of unserved or underserved. 23 22 See maps at www.cpuc.ca.gov/PUC/Telco/Information+for+providing+service/Broadband+Availability+Maps.htm. 23 The speed categories presented in the CPUC map (Figure 1) do not accord with the speeds that define unserved and underserved areas (Table 1), making it difficult to determine which areas qualify for federal stimulus funds based on the map. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 13 FIGURE 1 Broadband A vailability in California, August 2009 SOURCE: California Public Utilities Commission, Communications Division, Video Franchising and Broadband Deployment Group. NOTE: Map used with permission of CPUC. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 14 Limitations of Available Broadband Data FCC data collected prior to the December 2008 improvements represent the best public historical information on broadband availability, and policymakers and academics have relied on these data to analyze the extent of broadband deployment and the relationship between broadband and economic and social outcomes. 24 These historical data consist of the number of subscribers at the state level, disaggregated by advertised speed of service and type of technology (cab le, DSL, etc.), as well as a list of ZIP codes where the provider has at least one subscriber to “high -speed Internet service,” defined as at least 200 kbps in at least one direction. Thus, the only sub -state information was the number of providers in a ZI P code with at least one subscriber, which has been published in semiannual reports back to 1999. 25 Some caution must be exercised in interpreting these data, because m any ZIP code s cover large geographic areas, and providers with a subscriber in a ZIP code might not offer service throughout the entire ZIP code. That said, these ZIP code provider -count data have two strengths for assessing policies designed to raise broadband availability. First, broadband policies are often enabled by adding providers to an area — sometimes directly through public provision and sometimes indirectly throu gh subsidization or regulation; thus, using the number of providers as a proxy for broadband availability is meaningful in a policy context. Second, there is a reliable relatio nship between the number of providers in a ZIP code and the estimated extent of residential broadband availability , even though the data are imperfect. As explained in the technical a ppendix, the publicly available FCC data do not distinguish between providers that own the wires and cables that constitute broadband’s infrastructure and those that lease and resell space on these lines to offer service; nor do FCC data indicate how much of the ZIP code each provider serves, nor at what speed. Nevertheless, related research has shown that estimated broadband availability in a ZIP code does increase when the number of providers increases, especially when the ZIP code has fewer providers to begin with. 26 What We Know about Broadband Availability Broadband is widely available. By December 2006, essentially all ZIP Code Tabulation Areas (ZCTAs) in the United States, including those in rural areas, had at least one provider offering service ( Table 2). 27 24 National broadband data collected prior to December 2008 cannot show which areas qualify as unserved or underserved, as defin ed for ARRA. Historically, the FCC has reported the number of providers in a ZIP code with service at speeds of at least 200 kbps in at least one direction, whereas an area qualifies as unserved if it lacks service of 768 kbps downstream and 200 kbps upstream. Thus, area s with “high-speed access” according to the historical FCC definition could still be unserved or underserved according to definitions for curren t broadband programs. This simple metric overstates broadband availability because providers do not always offer service throughout an entire ZIP code. Using an improved methodology that infers availability based on adoption patterns and FCC provider counts, we estimate that broadband was available to 85 percent of U.S. households and 92 percent of California households in December 2005 (Kolko, forthcoming). Looking at the number of providers, broadband availability continued to increase between 1999 and 2006, relative t o its level in 1999. The average number of providers per 25 These semiannual reports and provider count lists are available at www.fcc.gov/wcb/iatd/comp.html . Two reporting changes occurred starting in December 2005: First, providers with fewer than 250 high -speed connections were required to submit data through Form 477, whereas previously they were exempt; se cond, providers had to start reporting state -level subscribership by technology (GAO, 2009). 26 Kolko (forthcoming) demonstrates the relationship between provider count and availability. This relationship suggests that th e logarithm of the number of provide rs better approximates availability than either a linear measure or an assumption that availability reaches a maximum with one or a few providers. The appendix of this report discusse s other shortcomings with FCC Form 477 data in more detail, as does GAO (2006). 27 ZCTAs are U.S. Census Bureau approximations of U.S. Postal Service ZIP codes. ZCTAs are better suited to data analysis than ZIP codes. See the appendix for more detail about how ZCTAs and ZIP codes compare and how we converted ZIP code data to ZCTAs. “Rural” means outside a metropolitan area and accounts for roughly 20 percent of the U.S. population. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 15 ZIP code in the nation grew from 3.4 in 1999 to 11.2 in 2006, weighted by employment. The comparable figures for California were slightly higher (5.0) in 1 999 and similar (11.5) in 2006. TABLE 2 Number of broadband providers by ZCTA All areas Rural only U.S. California U.S. California 1999 Average broadband provider count 1.6 2.9* 0.8 0.9 Average broadband provider count (employment -weighted) 3.4 5.0* 1.7 1.6 Percent of ZCTAs with one or more providers 63% 77%* 42% 43% Percent of ZCTAs with one or more providers (employment - weighted) 96% 98%* 83% 81% 2006 Average broadband provider count 6.9 8.6* 5.0 4.9 Average broadband provider count (employment -weighted) 11.2 11.5 8.3 7.4* Percent of ZCTAs with one or more providers 99.6% 99.93%* 99.4% 99.5% Percent of ZCTAs with one or more providers (employment- weighted) 99.97% >99.99%* 99.94% 99.99% NOTES: Asterisks indicate California value is different from rest of U.S. at 5% statistical significance level. “Rural” means not in a metropolitan area. See technical appendix for further definitions and details. Since broadband appears, by these data, to be widespread, we may ask why federal and California broad- band policy is focused on unserved and underserved areas . The answer is that ARRA, CASF, and other current initiati ves use higher speed thresholds than the historical FCC definition of high- speed access (see Table 1 ). It is very likely that many areas are unserved or underserved relative to these thresholds, even if they have one or more broadband providers according t o the historical FCC definition. Thus, even though historical data on broadband availability do not show which areas qualify as unserved or underserved for the purposes of current broadband initiatives, they do reveal that broadband became available at di fferent times throughout the country. These historical data allow us to identify areas with more and less broadband availability and whether broadband expansion was associated with economic development outcomes. Why There Is a Digital Divide in Broadband Availability Why might broadband be more widely available in some places than others? The primary reason is that the costs and benefits of providing broadband depend on local factors. Broadband provision requires fixed costs to extend service to an area: Mu ch of the cost to install or upgrade telecommunications infrastructure is required “up front,” regardless of the number of eventual subscribers served by that infrastructure. Thus, in order to spread the fixed costs across more subscribers, providers are m ore likely to serve areas with high demand for broadband. In addition, infrastructure is more expensive to deploy in some areas, such as those with steep terrain or fewer roads, as broadband lines often follow existing transportation rights -of -way. Finally , broadband availability can vary because most areas in the United States are served by a dominant telephone provider and a dominant cable television provider, and each can make different strategic decisions http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 16 about when to introduce broadband service to the ir regions. State policies about regulating or subsidizing broadband could also affect the level of availability. 28 As expected, ZCTAs with higher population density, higher income, and flatter terrain had more broadband providers in 2006, holding other factors constant ( Technical Appendix Table A1 ). Roughly half of the difference between the number of providers in California and the nation in 1999 ( see Table 2 ) was due to differences in density, income, education, and terrain; the remainder was due to unmeasured factors, which could include state policies or the particular broadband strategies of the telephone and cable companies serving California. By 2006, the gap in availability bet ween California and the United States had closed to a statistically insignificant difference. The extent and reasons for the digital divide in Internet availability also depend on technology. Compared with today’s main broadband technologies, cable and DSL dial -up Internet service involves lower fixed costs and did not lead to persistent geographic disparities in availability. Fiber -to -the -home, which offers speeds much faster that DSL or cable, has high fixed costs of deployment and is therefore likely to lead to a more persistent digital divide than cable and DSL (Kolko, 2007). Even if today’s bro adband technologies become available everywhere, digital divides in future technologies will probably appear, and “closing the digital divide” will remain a policy goal. 28 We do not attempt to measure the effect of state policies on broadband availability because telecommunications regulations are difficult to quantify in a consistent way across states. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 17 Research Questions and Approach In this chapter we first discuss the questions this report addresses, explaining the underlying hypotheses and why the questions are important. We then present the empirical strategy that we use in answering these questions. We conclude the chapter with a discussion of other important questions about broad band’s effects that lie beyond both the scope of our analysis and the main concerns of current federal and state policy initiatives. Questions Addressed in This Report Our analysis answers four questions about broadband expansion and economic growth: 1. Does employment grow faster in areas with greater broadband expansion? 2. Does the relationship between broadband and employment differ by industry or across places? For instance, is it stronger for industries that are more reliant on technology or that use worke rs who are more technically knowledgeable? Is it stronger in places that are more isolated or that have higher amenities? 3. If there is a positive relationship between broadband expansion and employment growth, does broadband expansion cause this growth? 4. If broadband does boost employment growth, who benefits? Is employment growth accompanied by a greater likelihood of employment, higher income, or greater flexibility to be able to work from home? To answer our first question—whether the availability of broa dband is associated with local employment growth —it is helpful to consider why broadband might contribute to this type of local economic development. Obviously, broadband and other information and communications technologies lower the cost of sending and receiving many forms of data, including documents and audio and video content. According to standard economic theory, lowering the cost of one input to a profit -maximizing firm has two possible effects: First, the firm raises its output; second, the firm sh ifts its mix of inputs toward the input whose cost went down. When the spread of broadband lowers the cost of communication, the net result of these two effects on hiring is theoretically ambiguous. The first effect —raising output —would lead most businesse s to hire more labor. But the second effect —shifting toward the inputs that just got cheaper —could cause businesses to use new technology in lieu of labor for some tasks. Our second question explores how the relationship between broadband and employment v aries across industries, types of workers, and places. Again, it is useful to think about why there might be differences. Turning to industries first: Businesses reliant on information technology might increase their employment of workers skilled in using new technology and possibly even reduce their employment of others —shifting, in other words, toward labor that is “complementary” with broadband technology. Broadband would therefore have a larger positive effect on employment in industries whose workers are more skilled in using information technology. Furthermore, even if broadband expansion caused individual firms to reduce employment, economic activity could shift to locations where broadband is more widely available, raising aggregate employment there relative to other areas. Locational shifts in economic activity would be more pronounced in more “footloose” industries —those whose location is not tied to local markets or inputs. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 18 The relationship between broadband expansion and economic growth could als o vary across places. Broadband might offer greater benefit for places that are smaller or more isolated, helping local businesses or households to connect with larger markets: This line of thinking lies behind predictions that rural areas might benefit di sproportionately from Internet technology. Other examples include areas that have access to a more highly educated labor market (if the more educated workers are better able to use advanced information technologies) and areas with more favorable climates a nd recreational opportunities (if broadband access allows firms to move further away from suppliers and customers toward locations appealing to employers and workers). 29 At the same time, broadband expansion could lead to declines in both employment and e conomic output in local businesses that begin to face competition from online businesses located elsewhere. For example, broadband expansion might encourage households to purchase goods online instead of in local stores, or to download movies rather than visit their local theaters. As with the other possible links between broadband and local economic activity, the effects will vary by industry: Those businesses most dependent upon local demand, such as retailers or entertainment, are more likely to be hurt by broadband expansion, depending on how easily their customers can switch from local businesses to their online counterparts. Turning to our third question, it is important to acknowledge that a positive empirical relationship between broadband expansion and economic growth does not, in itself, mean that broadband expansion causes economic growth. The reverse might actually be true if broadband providers choose to offer or expand service in areas that are growing faster. Alternatively, population growth c ould cause both broadband expansion and employment growth. Broadband expansion might follow population growth since more than 60 percent of broadband subscribers are households, according to the most recent FCC broadband report. Once again, this effect cou ld vary by industry: Population growth would lead to employment growth in industries whose customers are local residents. Assessing causality is, of course, essential for predicting whether broadband policies will lead to economic development. Looking at b roadband expansion and employment growth in individual industries might help clarify the relationship between broadband expansion and overall employment growth. Our final question asks who benefits if, indeed, broadband expansion does cause economic growth. The answer depends, in part, on what happens to the population. If population growth accompanies employment growth because people follow jobs, then the likelihood of residents being employed— the employment rate— might not rise much. The effect on average pay and household incomes is unclear as well: Increased labor demand might raise wages, but if labor supply also increases, this would push wages down. Employment growth that raises the employment rate, average pay, or household incomes benefits residents; employment growth that does not might still contribute to property owners’ land values or local governments’ tax base, benefitting some residents but not others. 30 It is quite plausible that targeting unserved and underserved areas for broadband expansion could raise local employment yet offer ambiguous economic returns for residents. Economists are often skeptical about “place -based” policies, which provide geographically targeted infrastructure investments; the broadband 29 “Why Wall Street Is Losing Out To 40 Acres and a Modem,” New York Times, December 27, 1998, highlights the relocation of some financial operations to idyllic recreation areas like Jackson Hole and Nantucket. 30 We do not assess the relationship between broadband expansion and property values because the time period we study (19 99–2006) coincides with the large and varying house price bubbles in many local housing markets. The geographic variation in price changes durin g this time is probably too noisy to be a reliable measure of the geographic variation in the capitalization int o land values of potential local productivity enhancements such as broadband expansion. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 19 stimulus funds are an example of s uch policies. According to Louis Winnick (1966), an economist and urban development expert, “Federal programs to change the geography of output are a kind of welfare device to redistribute personal income. But at best it is a clumsy, expensive, and often inequitable device. Not only are the gains to one locality offset by losses to another, but even in the locality of gain the added income frequently goes to the wrong people. … A disproportionately large share of the increased purchasing power goes to the owners of immobile resources [e.g. property owners] other than labor.” 31 Furthermore, place - based policies may end up encouraging economic activity in places where that activity might not otherwise be economically sustainable: Many of the places that are unserved and underserved by broadband are so because terrain, remoteness, or low population density raises the cost of broadband provision. 32 Research Strategy Yet, in defense of place- based policies, not all people can or want to leave disadvantaged places, and governments are committed to offering basic services to residents in all places. The relationship between broadband and economic outcomes at the local level is relevant for assessing policy that targets specific locations for broadband investments. As described above, broadband availability has varied across the nation, and this report takes advantage of these geographic differences to assess how the expansion of local broadband availability relates to changes in many economic outcomes. Several other studies have also looked at the relationship between broadband and economic development. They have generally found a positive relationship between broadband expansion and employment growth and a mixed relationship between broadband expansion and income (or wage) growth. By using richer data, better measurement, and a wider range of methods, this report offers more definitive answers about the relationship between broadband and economic growth, which are generally consistent with previous work. 33 Our research strategy consists of comparing changes in several economic outcomes wi th the extent of broadband expansion. To measure the expansion of broadband availability, we use the FCC’s Form 477 data on the number of broadband providers in a ZIP code, as described in the previous section and the This report also examines questions left unanswered in previous work, such as causality, the effect on population, and the effect on telecommuting and other forms of home -based work. technical appendix . We match these data with economic outcomes from several other data sources:  The change in employment provided in the National Establishment Time -Series (NETS) database, which reports employment for nea rly all businesses in the United States from 1992 to 2006 and includes detailed industry and geographic information. The NETS is based on the Dun & Bradstreet business register.  The change in employed residents, total and working -age population, average pa y, and median household income, all at the county level, from the U.S. Census and other government data sources.  The change in the likelihood of telecommuting, bringing work home, and operating a home -based business, as reported in surveys conducted by For rester Research, a technology consultancy. Forrester surveys households annually about technology adoption and related behaviors. 31 Ladd (1994), in a review of research on enterprise zones, also notes that the benefits of place -based policies to non -landowning local residents are uncertain. 32 One prominent urban economist (Edward Glaeser) argues that disaster- struck New Orleans (Glaeser, 2005) and economically depressed Buffalo (Glaeser, 2007) should be allowed to shrink, with support given directly to residents rather than to places. 33 The appendix offers more detail on related academic and policy research. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 20 The FCC, NETS, and Forrester all report information at the ZIP code level. ZIP code areas are small relative to counties and offer rich detail on local broadband availability and economic outcomes. But official ZIP codes, as defined by the U.S. Postal Service (USPS), change frequently over time and do not accord precisely with county boundaries, Census tracts, or other areas for which data are typically available. Furthermore, while the NETS and Forrester report actual USPS ZIP codes for businesses and households, respectively, the FCC uses its own approximation of ZIP codes. And the Census uses its own ZIP code approximation, the ZCTA (ZIP code tabulation area), for reporting selected demographic and economic measures. We converted data from all these sources to the Census ZCTA definitions to create consistent geographic boundaries across datasets and over time. Most previous research has examined the relationship between broadband and employment growth at larger geographic levels (e.g., counties or states). However, using smaller units of geography is more desirable because broadband availability can vary block -by -block. ZIP codes are the smallest geographic area for which both broadband and employment data are publicly available. 34 We focus on broadband expansion and outcomes between 1999 and 2006. Although broadband diffusion began earlier, most growth in the number of providers occurred during this period (the FCC began reporting provider count data in 1999). These years are also most relevant as a guide to what future broadband expansion—through ARRA or other programs —might mean for economic development. The relationship between broadband (or any technology) and economic outcomes might change over the course of the technology’s diffusion. 35 Policies designed to bring broadband to still -underserved areas are at the end of the broadband diffusion process (at current broadband speeds , anyway), so more recent experience is a better guide than earlier experience to what might happen in the future. 36 The first step, then, is to assess the overall relationship between broadband expansion and employment growth. We then look at whether the relationship varies by industry or type of place. Next, we try to assess ca usality by examining the industry -specific effects and by also using an “instrumental variable” strategy. To further refine these insights, our analysis highlights the relationship over the period 1999 –2006 in ZCTAs with the least broadband provi sion in 1999. Although data do not yet exist to assess whether the locations that lagged in 1999 are the same locations that remain unserved or underserved by today’s policy definitions, this analysis may offer guidance for the effects of policies targeting today’s unserved and underserved areas. 37 34 To assess the relationship between broadband expansion and employment growth, the ideal level of geography would be th e exact street address because a business can have broadband access only if it is available at its exact location. The ZCTA is the smallest geography for which data are available and is therefore the level of analysis we use for employment growth. In contrast, counties are more appropriate than ZCTAs for examining household labor market outcomes because counties better approximate the size of labor markets and broadband could affect households that are in the same labor market as businesses that adopt broad band. And finally, we examine other outcomes: population growth, changes in the likelihood of being employed, average pay, and household income, as well as the changing prevalence of three types of home -based work: telecommuting, bringing work home occasionally, and operating a home -based business. These additional outcomes reveal how broadband expansion relates to outcomes that households experience. 35 It is ambiguous whether the economic effects, theoretically, might be larger earlier or later in the diffusion process. Early users, by taking advantage of a technology first, might grow to a scale that later users would find difficult to compete with. However, the cost of adopting a new technology can fall over time with technological improvements and knowledge from lessons learned by earlier users, which migh t increase the economic benefits for later users. 36 Focusing on the 1999 –2006 period also allows us to adjust for earlier trends in employment growth, because the NETS data start in 1992. Adjusting for prior trends is important because it accounts for the possibility that earlier employment growth encouraged lat er broadband expansion, and late r employment growth might simply be the continuation of the earlier growth trend rather than the effect of broadband expansion. 37 The instrumental variable strategy identifies a factor – in this case, slope of terrain – that affects broadband expansion wi thout independently affecting employment growth, holding other factors constant. The relationship between employment growth and the variation in broadband expansion that is predicted by slope identifies the causal portion of the effect of broadband on grow th, at least for the areas in which slope is a good predictor of broadband expansion. The appendix offers detail on this approach and the results. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 21 Althou gh our analysis focuses on the 1999 –2006 period, there is also some value in looking at the 1992 –1999 period, when broadband first became available in some areas. 38 Questions This Report Does Not Address The relationship between broadband expansion and economic growth in this earlier period may offer guidance on how policies aimed at hastening the diffusion of early -stage technologies affect local economic development, assuming that next -generation technologies that fundamentally change business processes will have the same relationship to econom ic activity as did current broadband technology. Because federal and state broadband policies are particularly focused on unserved and underserved areas, this report looks at how local broadband expansion relates to l ocal economic outcomes. The methods used in this report and in related research examine the relationship between broadband expansion and local activity relative to other areas. Even if broadband had no net effect on employment growth nationally, we could s till see a positive relationship between broadband expansion and local employment growth if broadband expansion shifted economic activity from some locations to others, resulting in no net gains in jobs or output. If equity is an important public policy go al, governments could still favor broadband investments that redistribute economic activity toward underserved areas. Yet broadband policies could also affect economic activity nationally. Numerous countries have set goals not only to close the digital div ide by making a minimum broadband service available to all residents, but also to support higher -speed broadband services to keep high -tech industries globally competitive (Li and Losey, 2009). In its own evaluation of broadband deployment in the United St ates, the FCC considers both the extent and evenness of broadband access among Americans —the digital divide—as well as comparisons with other countries, though there is disagreement about which measures accurately reflect international differences in broadband availability and, accordingly, disagreement about the international ranking of the United States. 39 Broadband may also offer benefits not ful ly captured in measures of output, employment, or growth. People use broadband for a wide range of activities, including many —such as sharing pictures with friends or downloading music—that might not fulfill a public policy goal, even though people value t hese activities. Other benefits —for example, access to news, remote medical services, or distance learning —might be public policy goals but might not be reflected in standard economic indicators such as employment, output, or income. Thus, looking at the r elationship between broadband and economic outcomes might exclude some benefits of broadband that do achieve public policy goals and others that do not. Although our analysis in this report focuses on local economic development measures, we also review the limited recent research on other outcomes in our final chapter. It is thus challenging to estimate the net national benefits of broadband, and attempting to do so requires different methods. Although this subject lies beyond the scope of our report, we do return to the question of national benefits briefly in the conclusion, as measured by consumer willingness to pay for broadband. 38 As the appendix explains, a methodological disadvantage of looking at the 1992 –1999 period is that we have to assume that no ZIP code had any broadband providers in 1992, so the level of provi ders that the FCC reported in 1999 equals the change between 1992 and 1999. Another disadvantage of looking at the 1992 –1999 period is that we cannot adjust for prior trends in employment because the NETS data start in 1992. 39 FCC 08 -88, Fifth Report on br oadband deployment, as required by Section 706 of the 1996 Telecommunications Act, June 2008. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 22 Research Findings Broadband and Local Overall Employment Growth The overall relationship between broadband expansion and employment growth, as measured by the NETS, is positive. Moving from no broadband providers to 1 –3 providers (the FCC groups one, two, and three providers together in its reporting) is associated with employment growth that is higher by 6.4 percentage points over the seven -year period from 1999 to 2006 (Table 3). 40 The size of this relationship is large relative to the overall national employment growth rate, but employment growth at the ZCTA level shows wide variation, and broadband expansion accounts for relatively little of the variation in growth rates across ZCTAs. 41 technical appendix tables The relationship between broadband expansion and employment growth is similar when looking only at areas with few or no broadband providers in 1999. As explained above, these less -well -served ZCTAs in 1999 are the best guide to the relationship between future broadband expansion and employment growth in the areas targeted today by federal and state programs to raise broadband availability. Broadband expansion is associated with a smaller but still statistically significant increase in employment growth of 3.5 percentage points in the 1992 –1999 period. (Many of the results discussed in this section are shown in the .) TABLE 3 Broadb and and economic outcomes , 1999–2006 Percentage point change associated with increase in broadband availability Employment growth (ZCTA) 6.4* Working age population (county) 2.4* Employed residents / working age population rate (county) -1.2 Average pay per employee (county) 1.1 Median household income (county) -2.4* NOTES: Percentage changes reflect the coefficient of the change in the dependent variable, in lo g form, for a shift from 0 to 1– 3 providers or from 1– 3 to 4 providers, or equivalent changes in the log of the number of providers. Asterisks denote significance at the 5% level. Employment growth is from the NETS; remaining variables are from Census and BLS. See technical appendix for detailed explanation and complet e results. 40 Our regression analysis uses the change in broadband providers in logarithmic form, so adding more providers has a less -than -proportional additional effect on employment and other outcomes. See appendix for details. 41 The cumulative national employment growth rate in 1999 –2006, according to the NETS, was 0.1 percent, which is well below Census estimates based on business surveys (6.2%) and household surveys (7.5%). The NETS is known to undercount some employment in the most recent years, though local employment growth across counties is highly correlated between the NETS and Census business surveys, especially when looking over multiyear employment changes rather than one -year changes. See Kolko and Neumark (2007) for d iscussion of the NETS and its comparison to other data sources. The standardized beta of the relationship is .08, meaning that a one standard -deviation increase in broadband availability is associated with a .08 standard deviation increase in employment growth. See appendix for details. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 23 Broadband and Employment Growth a cross Industries and Places The relationship between broadband expansion and employment growth varies across industries. The positive relationship is especially large for utilities; information; finance and insurance; professional, scientific, and technical services; management of companies and enterprises; and administrative and business support services. The relationship in these sectors is much larger than the relationship for overall employment (Table 4) and is positive and statistically significant for all but two sectors: mining and public administration. TABLE 4 Broadband and industry employment growth, 1999–2006 Employment growth Percentage point employment change associated with increase in broadband availability Highest share of technology inputs Highest share in computer occupations Employment location most tied to population Management of companies and enterprises (55) 40.8*** X X Utilities (22) 16.7*** X Professional, scientific, and technical services (54) 16.4*** X X Finance and insurance (52) 14.8*** X Administrative and business support services (56) 14.1*** X Information (51) 12.0*** X X Construction (23) 11.8*** X Agricultural, forestry, fishing, and hunting (11) 11.6*** Real estate and rental and leasing (53) 10.2*** X Accommodation and food services (72) 9.9*** Transportation and warehousing (48-49) 8.6*** Health care and social assistance (62) 7.4*** Wholesale trade (42) 7.1*** Other services (81) 7.1*** X Mining (21) 6.6 Retail trade (44-45) 6.5*** X Manufacturing (31-33) 6.3** Educational services (61) 6.1*** X X Arts, entertainment, and recreation (71) 5.7** Public administration (92) 0.5 NOTES: Numbers in parentheses are the NAICS codes for the industry sector. Industries are ranked by the employment change associated with in creased broadband availability. Percentage changes reflect the coefficient of the change in the dependent variable, in log form, for a shift from 0 to 1 –3 providers or from 1 –3 to 4 providers, or equivalent changes in the log of the number of providers. Asterisks denote significance at the 5% level. Employ ment growth is from the NETS. See technical appendix for detailed explanation and complete results. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 24 The relationship between broadband and employment tends to be stronger in industries where information technology (IT) services (Internet publishing, telecommunications services, data processing, and related services) represent a larger share of an industry’s inputs. These industries include: information; professional, scientific, and te chnical services; management; administrative services; and educational services. Of these, all but educational services are among the industries whose employment growth shows the strongest relationship with broadband expansion. In addition, industries wit h a larger share of employees in computer specialist occupations tend to show a stronger relationship between broadband expansion and employment growth. Utilities; information; finance and insurance; professional, scientific, and technical services; and ma nagement had a higher share of employees in these occupations, and all five of these industries are among those showing the strongest relationship between broadband expansion and employment growth. 42 A possibility raised above is that some businesses could be hurt by broadband expansion if online services compete with traditional businesses. Retail is one example, if people with broadband access choose to shop online instead of at local retailers. Another could be the arts, entertainment, and recreation industry, whic h could see less local demand for live or on -site events if broadband makes online substitutes available. Both of these industries show weaker relationships between broadband expansion and employment growth, relative to other industries. In sum, industries that rely more on technology inputs an d on workers in computer specialist occupations— the industries that should benefit more from broadband —are those in which broadband expansion is associated with stronger employment growth. The relationship between broadband and employment growth is also stronger in some places than others. For example, the relationship is stronger for ZCTAs with lower population density (and, conversely, weaker for those with higher density) —consistent with the theory that s maller or more isolated areas may benefit more from high -speed connections, giving businesses in these areas access to larger markets. However, even for most high density areas, the relationship between broadband and growth remains positive on balance, jus t not as large as for lower -density areas. None of the other place characteristics —such as having a more educated workforce, having a better climate, or being a vacation destination —affects the relationship between broadband expansion and employment growth ( Technical Appendix Table A4 ). Does Broadband Expansion Cause Employment Growth? A crucial question for broadband policy is whether broadband expansion causes employment growth: If so, then policies to make broadband available should result in local growth. Conceivably, a positive relationship between broadband expansion and employment growth could arise for other reasons. For example, if broadband providers ex pand in locations where they anticipate future growth, then the positive relationship would in part or entirely reflect this strategic decision of providers rather than a causal effect of broadband on growth. Alternatively, population growth could cause both broadband expansion and employment growth: Broadband providers could invest in areas where population (and therefore demand for broadband) is growing, while at the same time population growth could cause employment growth in industries (such as retail, restaurants, and personal services) that serve local populations. 42 The appendix explains these two industry -level measures of technology intens ity. “Inputs” are based on the Bureau of Economic Analysis’s input -output tables, which describe the types of inputs – including IT services –each industry uses in producing its output. Occupational data, by industry, are from the Bureau of Labor Statistics. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 25 Our results seem to rule out the idea that broadband providers are explicitly targeting areas where they expect higher economic growth: Later employment growth does not predict earlier broad band growth. The evidence also indicates that population growth is not the main driver. Adjusting the analysis for population growth changes the relationship between broadband expansion and employment growth only minimally: The boost to employment growth f alls from 6.4 to 5.0 percentage points but remains statistically significant ( Technical Appendix Table A2 ). Furthermore, the sectors whose employment is most tied to population are not those that show the strongest relationship between broadband expansion and employment growth. 43 A more technical assessment of causality uses the instrumental variable approach described above, in which slope of terrain in a ZCTA is used as prox y for broadband expansion that should not have any independent relationship with employment growth. Jobs tend to locate where people do in industries such as construction, real estate rental and leasing, education, retail trade, and other personal services; t hese industries are not among those in which employment grows most where broadband expands ( see Table 3). 44 Technical Appendix Table A5 Using this approach, the relationship between broadband expansion and employment growth remains positive and statistically significant, suggesting a causal relationship. However, using this approach, the size of the relationship appears implausibly large and is quite sensitive to how the regression model is specified, which makes us less confident in this result ( ). This approach, therefore, is only modestly suggestive that broadband expansion causes employment growth. These different approaches, though not definitive, generally point in the direction of a caus al relationship and suggest that broadband expansion leads to large increase in local employment growth. Broadband and Household Outcomes To assess how the employment growth associated with broadband expansion benefits households, we consider key labor market outcomes. First, we look at the employment rate: the percentage of working -age residents that are employed. Other important outcomes are average pay per employee and median household income, which could change with employment growth because of changes in labor demand, labor supply, or the composition of the workforce. 45 Broadband expansion has no statistically significant relationship with the employment rate. As discussed above, it does not appear that population growth causes both broadband expansion and employment growth. Rather, people may be following jobs: The employment growth associated with broadband expansion could encourage people to move to areas where employment opportunities are expanding. Furthermore, although most people work in their county of residence, some people commute across county lines. In counties where many employees live elsewhere or where many residents work elsewhere, broadband expansion is asso ciated with an increase in employment but not in the number of employed Finally, we consider whether broadband availability facilitates working from home. 43 The appendix explains this industry -level measure of whether employment is tied to population, which is based on the similarity between the geographic distribution of employment in an industry and the geographic distribution of population. 44 As the appendix explains, the regression includes variables such as road density that, if omitted, could lead slope and employment growth to have a direct relationship and would invalidate this approach. 45 The analyses of household outcomes look at the county, rather than the ZCTA, level. Although businesses should be directly af fected by broadband expansion only at their e xact address, increased demand for employment could benefit workers throughout a labor market, and counties —which are almost always larger than ZCTAs —are a better approximation of labor markets than ZCTAs. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 26 residents. In other words, some new jobs associated with broadband expansion aren’t filled by local residents: Employment opportunities attract people who are willing to move or commut e. 46 Broadband expansion is associated with no change in average pay per employee and a decrease in median household income. The results for counties with less broadband availability in 1999 —the group which may be most comparable to the places that current broadband policy targets—are similar: There is no relationship between broadband expansion and either the employment rate or average pay per employee. In sum, whatever positive effects broadband may have on employment growth, it did not result in either h igher employment rates or higher pay for residents in areas where broadband expanded during the period 1999 – 2006 period. As described above, these residents may benefit indirectly from employment growth in their county if that growth raises the local tax b ase or property values, but they do not benefit directly in terms of greater likelihood of being employed or higher earnings or income. Residents who rent, and therefore might have to pay more for housing when property values rise, could possibly become worse off economically from employment growth that raises neither the employment rate nor average pay per employee. The household outcomes reveal a difference between the earlier and later time periods of broadband’s diffusion. In contrast to the later period, broadband expansion from 1992 to 1999 showed positive and statistically significant relationships with average pay per employee, median household income, and the employment rate ( Technical Appendix Table A6 ). One possible explanation is that businesses that adopted broadband earlier faced a labor market where computer skills were less widespread than in later years; these early adopters might also have had to hire workers with more advanced skills if they had to develop more applications in -house to integrate a nascent and fundamentally new technology into their business processes. Later adopters could rely more on off -the -shelf mass -market applications that workers with more modest technology skills could use. 47 Finally, we examine the link between expanded broadband availability and the likelihood of working at home. At -home work could benefit households by giving them (and their employers, in the case of telecommuting) more flexibility. We find no relationship between broadband expansion and all three types of home -based work: (i) having a formal relationship with an employer to work at home at least one day a week (which the survey refers to explicitly as “telecommuting”); (ii) bringing work home to do outside of normal work hours; and (iii) operating a business from home ( If the supply of workers with the appropriate technology skills were more limited in the earlier period, that could result in both higher increases in the employment rate and in average pay per employee in areas with greater b roadband expansion. Technical Appendix Table A7 ). Even for the types of people whose jobs or skills might lend themselves more easily to home-based work —people with college degrees or in managerial or professional occupations —there was no relationship between broadband expansion and changes in doing home -based work. The absence of a relationship between broadband expansion and home -based work may seem surprising, yet recent experience shows that telecommuting —even in places where technology makes telecommuting possible —is rarer than some have expected. One reason for this might be that communication with remote work ers through “telepresence” or other video communications often involves expensive hardware and 46 Note that the number of employed residents and the employment rate (reported by the Census) are based on whether residents of a county are employed, regardless of where their employer is located. In contrast, employment growth (reported by the NETS) is based on wh ere businesses are located, regardless of where their employees live. See appendix for further discussion. 47 The NETS data provide information on the number of employees in a business but no information about their education or skill level. Data on the characteristics of workers in firms adopting broadband technology at different stages of broadband diffusion would help a ssess these conjectural explanations. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 27 much faster broadband connections than the minimum speeds that qualify as “broadband.” Recall that the speed necessary for telecommuting according to the CPUC —3 mbps downstream—is fifteen times faster than what qualifies as broadband in the historical FCC definition. Another plausible explanation is that corporate culture can deter telecommuting, even if all of the requisite technology is available: Telecommuters often feel invisible to their employers, and managers often feel a loss of control and uncertainty about the output of remote workers. 48 The relationship between broadband and telecommuting could change, however, with the development of low -cost, high- quality video technology and management styles more conducive to remote work. 48 “Home Warriors: Telecommuters Need More Than E- mail and a Broadband Connection,” The Economist , July 25, 2008. “Top 10 Myths About Mobile Work,” Sun Microsystems Open Work Services Group White Paper, 2008. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 28 Implications and Conclusion What the Results Mean for Broadband as Local Economic Development Policy The findings indicate a positive relationship between broadband expansion and em ployment growth (Table 5). The evidence suggests that broadband may have a causal effect on employment growth: The relationship is strongest in more technology -reliant industries, and population growth does not appear to be the trigger, which was the most plausible alternative explanation to a causal relationship because most broadband subscribers are households, not businesses. TABLE 5 Summary of f indings Economic outcome Relationship with broadband Possible reason Employment Positive and big Broadband expansion causes existing businesses to expand or redistributes economic activity toward the area Working-age population Positive People are mobile and move to where employment opportunities are expanding Employment rate (employed residents / working-age population) None Because people are mobile or willing to commute, l abor supply grows along with labor demand, and businesses need not pay a large premium for basic technology skills Average pay per employee None Median household income Negative Telecommuting None Home-based work requires faster speeds than the minimum that qualifies as broadband, and many corporate cultures are not well-suited to telecommuting Bringing work home None Having home-based business None NOTE: Employment data are from NETS; population, employed residents, employment rate, average pay, and median household income data are from Census/BLS; telecommuting, bringing work home, and having home- based business data are from Forrester. However, the large increase in employm ent growth associated with broadband expansion does not necessarily benefit local residents. Areas with faster broadband expansion between 1999 and 2006 experienced no greater increases in either the employment rate—employment as a share of the working -age population —or in average pay per employee, relative to other areas, and median household income declined. One possible explanation for this is that broadband does indeed lead to employment growth, which encourages people to move or commute to areas where employment opportunities have expanded, and this increase in the local labor supply prevents the increased demand for labor from raising either the employment rate or average pay. Employment growth might still raise local property values and tax bases, but in the absence of more direct benefits for residents, the economic development benefits of broadband are ambiguous. As discussed above, place -based policies —for example, targeted broadband infrastructure investments —often involve uncertain economic benefi ts for residents. This debate surrounding place -based policies such as broadband infrastructure investment, along with our finding that broadband expansion raises neither the employment rate nor the average pay of residents, begs the question of whether pu blic money designated for broadband infrastructure might have a larger effect on http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 29 economic or social outcomes if the funds were allocated instead toward subsidizing broadband adoption or other needs of disadvantaged households, regardless of where they live . There is no easy answer; but it depends to some extent on the contribution of other potential benefits of broadband, which we discuss below, and how much society values investment in disadvantaged places. One caution with regard to our findings is that the best publicly available data on broadband’s diffusion — the FCC’s count of providers in a ZIP code —is an imperfect measure of availability. Although the provider count measure we use in our study is a good proxy for availability and an improvement over previous studies, it does not take broadband speed into account, which can vary considerably across different locations. This limitation may be especially relevant for our finding of no association between broadband expansion and any of three types of home -based work, including telecommuting. While broadband service that meets the minimum speed in FCC’s historical definition of broadband may not raise the prevalence of telecommuting, future broadband services t hat support videoconferencing and other “telepresence” applications might do so. Better broadband data will make the relationship between broadband speed and economic outcomes clearer in future research. Improving Data for Measuring and Evaluating Broadband The federal stimulus includes significant funds for broadband mapping and data collection, and the National Broadband Plan has the potential to propose additional forms of broadband research. We recommend two research directions to further the understanding of broadband and its effects. First, we should not only measure the availability, adoption, and use of broadband but also identify potential benefits of broadband and assess whether broadband diffusion is meet ing those goals. ARRA is focused on measuring access to broadband but not its effects. Second, although the FCC has already begun to collect better data on broadband availability since December 2008, earlier data on broadband availability —including the data we use in this report —could be much improved with better geographic specificity and information about speeds. To the extent that the FCC or other agencies have collected these data in the past but have not made them publicly available, they should do so. With a better understanding of where exactly and at what speeds broadband became available, researchers can more accurately assess the relationship between broadband and economic development and other outcomes —and offer better guidance to policymakers abo ut how to maximize the benefits of future broadband policies. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 30 Other Possible Benefits of Broadband Although economic development is a primary goal of ARRA and other initiatives seeking to raise broadband availability, ARRA, in the National Broadband Plan, directs the FCC to consider other possible benefits of broadband, including consumer welfare, health care, education, civic participation, government services, energy independence and eff iciency, and public safety. 49 Very few studies have examined the effect of broadband on these other outcomes. 50 The research on consumer welfare tries to place a dollar value on broadband access, based on what consumers are willing to pay. Yet estimates vary wildly beca use of different assumptions about how many residents value their home broadband connections well above the $30 –50 per month that many households pay, and how much these people would actually be willing to pay. The estimates of the consumer benefit of broa dband range from a few billion dollars per year to hundreds of billions per year, with little way to assess which assumptions about consumer willingness to pay are more accurate. The most compelling research in these areas either estimates the consumer welfare of broadband based on demonstrated willingness to pay for servi ce or considers how switching from dial -up to broadband Internet access changes online behaviors. 51 Recent research shows that people switching from dial-up to broadband spend more time using the Internet but that only some online behaviors change (Kolko, 2009). Yet even if it were possible to pin this number down, people might value bro adband for activities without any clear public benefit, such as downloading music, rather than for activities often believed to be in the public interest, such as increasing civic participation. To answer the policy question of why government should suppor t broadband, the fact that consumers want broadband is not enough; consumers value many goods and services, such as cars and cable television, which governments do not subsidize. The case for government support of broadband depends on what consumers use broadband for and what effects broadband has. 52 Other research has identified improved health care as a realistic benefit of broadband. One study estimates that broadband will save $30 billion per year in reduced medical costs and reduced institutionalization due to electronic medical records, remote monitoring of health indicators, and other health applications that depend on broadband (Litan, 2005). Policies have already begun to reflect these possible benefits of Switching to broadband increases the likelihood and intensity of several activities that do not fall under goals promoted by broadband policy: downloading music, purchasing products online, and visiting adult entertainment sites. Among “socially desirable” online activities that do fall under these goal s, the only activity that increases is researching health information. There is no change in visits to job or career websites or government websites, including civic participation activities such as getting information about public hearings or contacting e lected officials. Changing online behaviors will not necessarily lead to better social outcomes. For example, researching health information online will not necessarily enable people to live longer or healthier lives. But it is difficult to see how broadba nd will come to affect policy goals such as civic participation without changing online behaviors. 49The FCC’s Notice of Inquiry asks whether broadband could contribute to energy independence and efficiency through teleworking, a “smart” electricity grid, or “intelligent” highways that monitor traffic. Fuhr and Pociask (2007) suggest numerous ways that broadban d and other information technologies could achieve environmental and energy goals, including better supply -chain management, fewer printed materials, and reductions in driving and business travel. 50 Some research has looked at the relationship between Internet access in general and outcomes such as community participation and social ties, racial discrimination in retail transactions, book prices, and other outcomes, but not at broadband specifically. Since dial- up access has long been available throughout the United States, the relevant question for policy is the effect of having broadband access relative to ha ving dial- up access. Studies looking at the effect of Internet access generally do not help us understand the marginal effect of broadband. 51 See the appendix for more detail on willingness -to -pay estimates by Crandall, Jackson, and Singer (2003) and Greenstein and McDevitt (2009). 52 Other studies, including Anderson (2008) and Hitt and Tambe (2007), also found that broadband adopters increase their time on line, but the researchers’ me thodology did not allow them to identify effects on specific online activities. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 31 broadband on health. CETF’s largest award to date has been to the California Telehealth Network (CTN), connecting clinics, hospitals, and medical centers throughout the state. At the federal level, the FCC’s Rural Health Care Pilot Program began in 2007 to allocate over $400 million for broadband infrastructure and design for health care purposes; this program also supports CTN. Although we point to health care as a potential area of benefit, the evidence so far is based primarily on online behavi ors. Further studies are needed to assess the relationship between broadband availability and social outcomes, even if these effects are too long -term to be immediately measurable: health outcomes could include illness incidence or mortality rates; educati on outcomes could include attainment of a high school diploma or college degree. If future studies show evidence of broadband’s benefit for health or other social outcomes, broadband policies could consider weighing multiple criteria in determining which g eographic areas might benefit the most from closing the digital divide. For instance, policies could explicitly take into account not only which areas have the least broadband availability, but also which areas have the most pressing need for improvements in health care or another outcome that future research discovers is improved by increasing access to broadband. Broadband: The Next Generation Federal and state broadband policies seek to invest in broadband in underserved areas to close the digital divide. Yet closing the digital divide is not the only broadband policy that could affect economic development outcomes; and policies that support and hasten early -stage rollout of next -generation technologies, such as extremely fast fiber- to-the -home or new te chnologies that affect business processes, could have different economic development outcomes than those we report. What does appear certain is that even if the United States achieves the policy goal of ubiquitous broadband availability at current speeds, and the current broadband digital divide closes, new digital divides will open. Fiber -to -the -home, for example, is likely to promote even more extreme geographic disparity in availability than current broadband because the fixed costs of provision are so high (Kolko, 2007). Recent maps of broadband availability in California already show disparities across the state in available speeds. As average speeds increase, new applications develop that take advantage of —and require—these higher speeds, making what may have been an adequate broadband connection in the past inadequate for applications deemed in the future to be important for participating in the economy or society. Federal and state governments might again consider policies to increase the availabilit y of new technologies in less-well -served areas, expecting specific economic or social benefits from the expansion of the new technology. Our research on recent broadband expansion shows that technologies that contribute to local employment growth might no t, in fact, benefit local residents; and we must carefully consider who would ultimately benefit from public investments in next -generation broadband and other new technologies. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 32 References Anderson, Ben. 2008. “The Social Impact of Broadband Household In ternet Access.” Information, Communication, and Society 11 (1): 5 –24. California Public Utilities Commission . 2006. ConnectingCalifornia: Broadband Report Update. Available at ftp://ftp.cpuc.ca.gov/PUC/Telco/Reports/california+broadband+report+for+sept+2006+cetf+meeting.pdf Crandall, Robert, Charles Jackson, and Hal Singer. 2003. “The Effects of Ubiquitous Broadband Adoption on Investment, Jobs, and the U.S. Economy.” Criterion Economics, L.L.C. Released by New Millennium Research Council. Available at http://newmillenniumresearch.org/archive/bbstudyreport_091703.pdf. Crandall, Robert, William Lehr, and Robert Litan. 2007. “The Effects of Broadband Deployment on Output and Employment: A Cross -sectional Analysis of U.S. Data. ” Issues in Economic Policy discussion paper no. 6. Brookings Institution. Available at http://www.brookings.edu/~/media/Files/rc/papers/2007/06labor_crandall/06labor_crandall.pdf. Federal Communications Commission . 2006. “High -Speed Services for Internet Access: Status as of December 31, 2005 .” Available at http://fjallfoss.fcc.gov/edocs_public/attachmatch/DOC -266596A1.pdf. Flamm, Kenneth . 2009. “Diagnosing t he Disconnected: Where and Why I s Broadband Access Unavailable in the U.S.?” Unpublished manuscript . Forman, Chris, Avi Goldfarb, and Shane Greenstein. 2009. “The Internet and Local Wages: Convergence or Divergence?” NBER Working Paper #14750 . Fuhr, Joseph, and Ste phen Pociask. 2007. “Broadband Services: Economic and Environmental Benefits.” American Consumer Institute. Gillett, Sharon, and William Lehr. 1999. “Availability of Broadband Internet Access: Empirical Evidenc e.” Unpublished manuscript . Gillett, Sharon, W illiam Lehr, Carlos Osorio, and Marvin Sirbu. 2006. “Measuring the Economic Impact of Broadband Deployment.” U.S. Department of Commerce, Economic Development Administration . Glaeser, Edward. 2005. “ Should the Government Rebuild New Orleans, or J ust Give Residents Checks?” The Economists’ Voice 2 (4): article 4. Glaeser, E dward. 2007. “Can Buffalo Ever Come Back? ” City Journal 17 (4). Available at http://www.city - journal.org/html/17_4_buffalo_ny.html . Glaeser, Edward, Hedi Kallal, Jose Scheinkman, and Andrei Shleifer . 1992. “Growth in Cities.” Journal o f Political Economy 100 (6). Government Accountability Office. 2006. Broadband Deployment Is Exten sive throughout the United States, but It Is Difficult to Assess the Extent of Deployment Gaps in Rural Areas. GAO -06-426. Report to congressional committees . Government Accountability Office. 2009. Current Broadband Measures Have Limitations, and New Measures Are Promising but Need Improvement. GAO -10-49. Report to congressional committees. Greenstein, Shane, and Ryan McDevitt. 2009. “The Broadband Bonus: Accounting for Broadband Internet’s Impact on U.S. GDP.” NBER Working Paper 14758 . Grubesic, Tony . 2003. “Inequit ies in the Broadband Revolution. ” Annals of Regional Science 37:263–89. Grubesic, Tony. 2006. “A Spatial Taxonomy of Broadband Regions in the United States.” Information Economics and Policy 18. Grubesic, Tony, and Alan Murray. 2004. “Waiting fo r Broadband: Local Competition and the Spatial Distribution of Advanced Telecommunication Services in the United States.” Growth and Change 35 (2). Gunkel, David. 2003. “Second Thoughts: Toward a Critique of the Digital Divide .” New Media & Society 5 (4): 499–522. Henderson, Vernon, Ari Kuncoro, and Matt Turner. 1995. “Industrial Development in Cities. ” Journal of Political Economy 103 ( 5) . Hitt, Loren, and Prasanna Tambe. 2007. “Broadband Adoption and Content Consumption.” Information Economics and Policy 19 (3–4): 362–78. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 33 Kolko, Jed. 2000. “The D eath of Cities? The Death of Distance? Evidence from the Geography of Commercial Internet Usage.” In The Internet Upheaval: Raising Questions, Seeking Answers in Communications Policy , e d. Ingo Vogelsang and Benja min M. Compaine (Cambridge, MA: MIT Press), 73 –98. Kolko, Jed . 2007. “Broadband for All? Gaps in California’s Broadband Adoption and Availability .” California Economic Policy 3 (2). Available at http://www.ppic.org/content/pubs/cep/EP_707JKEP.pdf . Kolko, Jed, In press. “How Broadband Changes Online and Offline Behaviors.” Information Economics and Policy. Kolko, Jed. Forthcoming . “A New Measure of U.S. Residential Broadband Availability.” Telecommunications Policy . Kolko, Jed, and David Neumark. 2007. Business Location Decisions and Employment Dynamics in California. San Francisco: Public Poli cy Institute of California . Ladd, H elen. 1994. “Spatially Targeted Economic Development Strategies: Do They Work?” Cityscape: A Journal of Policy Development and Research 1 (1): 193−218. Li, Chiehyu, and James Losey. 2009. “100 Megabits or Bust! An Overview of Successful National Broadband Goals from Around the Globe.” Policy paper, New Americ a Foundation. Available at http://www.newamerica.net/publications/policy/100_megabits_or_bust . Litan, Robert. 2005. “Great Expectations: Potential Economic Benefits to the Nat ion from Accelerated Broadband Deployment to Older Americans and Americans with Disabilities. ” New Millennium Research Council . Available at http://newmillenniumresearch.org/arc hive/Litan_FINAL_120805.pdf. Osorio, Carlos. 2006. “The Economi c Impact of Municipal Broadband.” Unpublished manuscript . Prieger, James. 2003. “The Supply Side of the Digital Divide: Is There Equal Availability in the Bro adband Internet Access Market?” Economic Inquiry 41 (2). Shideler, David, Narine Badasyan, and Laura Taylor . 2007. “The Economic Impact of B roadband Deployment in Kentucky. ” Regional Economic Development 3 (2): 88–118. Song, Moohoun, Peter Orazem, and Rajesh Singh. “Broadband Access, Telecommuting, and the Urban -Rural Digital Divide. ” Iowa State Department of Economics Working Paper #06002. Stenberg, Peter, Mitchell Morehart, Stephen Vogel, John Cromartie, Vince Breneman, and Dennis Brown. 2009. “Broadband Internet’s Value for Rural Am erica.” U.S. Department of Agriculture, Economic Research Service. Van Gaasbeck, Kristin, Stephen Perez, Ryan Sharp, Helen Schaubmayer, Angela Owens, and Lindsay Cox. 2007. Economic Effects of Increased Broadband Usage in California. Summary report, Sacram ento Regional Research Institute. Wallsten, Scott. 2009. “Understanding International Broadband Comparisons: 2009 Update.” Technology Policy Institute. Winnick, Louis. 1966. “Place Prosperity vs. People Prosperity: Welfare Considerations in the Geographic Redistribution of Economic Activity.” In Essays in Urban Land Economics , by James Gillies et al. (Los Angeles: University of California Real Estate Research Program). http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 34 About the Author Jed Kolko is an associate director of research at the Public Policy Institute of California, responsible for managing the institute's economy research. He has conducted numerous studies of the California economy, economic development, housing, and technology policy. Prior to coming to PPIC in 2006, he was vice president an d research director at Forrester Research, a technology consultancy, where he managed the company’s consumer market research businesses and served as the lead researcher on consumer devices and access technologies. Jed has also worked at the Office of Fede ral Housing Enterprise Oversight, the World Bank, and the Progressive Policy Institute. He holds a Ph.D. in Economics from Harvard University. Acknowledgements I thank Gary Bjork, Michael Byrne, Ellen Hanak, William Lehr, David Neumark, Anne Neville, Micha el Teitz, and Lynette Ubois for thoughtful reviews, suggestions, and insights. I am grateful to Davin Reed for excellent research support. Participants at the 2008 Telecommunications Policy Research Conference offered helpful suggestions on an earlier vers ion, and Forrester Research graciously gave me permission to use its Technographics survey data. PUBLIC POLICY INSTITUTE OF CALIFORNIA Board of Directors Walter B. Hewlett , Chair Director Center for Computer Assisted Research in the Humanities Mark Baldassare President and CEO Public Policy Institute of California Ruben Barrales President and CEO San Diego Chamber of Commerce John E. Bryson Retired Chairman and CEO Edison International Gary K. Hart Former State Senator and Secretary of Education State of California Robert M. Hertzberg Partner Mayer Brown LLP Donna Lucas Chief Executive Officer Lucas Pub lic Affairs David Mas Masumoto Author and farmer Steven A. Merksamer Senior Partner Nielsen, Merksamer, Parrinello, Mueller & Naylor, LLP Constance L. Rice Co- Director The Advancement Project Thomas C. Sutton Retired Chairman and CEO Pacific Life Insura nce Company Carol Whiteside President Emeritus Great Valley Center The Public Policy Institute of California is dedicated to informing and improving public policy in California through independent, objective, nonpartisan research on major economic, social, and political issues. The institute’s goal is to raise public awareness and to give elected representatives and other decisionmakers a more informed basis for developing policies and programs. The institute’s research focuses on the underlying forces shaping California’s future, cutting across a wide range of public poli cy concerns, including economic development, education, environment and resources, governance, population, public finance, and social and health policy. PPIC is a private operating foundation. It does not take or support positions on any ballot measures or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. PPIC was established in 1994 with an endowment from William R. Hewlett. Mark Baldassare is President and Chief Executive Officer of PPIC. Walter B. Hewlett is Chair of the Board of Directors. Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright n otice is included. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Copyright © 2010 Public Policy In stitute of California All rights reserved. San Francisco, CA PUBLIC POLICY INSTITUTE OF CALIFORNIA 500 Washington Street, Suite 600 San Francisco, California 94111 phone: 415.291.4400 fax: 415.291.4401 www.ppic.org PPIC SACRAMENTO CENT ER Senator Office Building 1121 L Street, Suite 801 Sacramento, California 95814 phone: 916.440.1120 fax: 916.440.1121" } ["___content":protected]=> string(102) "

R 110JKR

" ["_permalink":protected]=> string(90) "https://www.ppic.org/publication/does-broadband-boost-local-economic-development/r_110jkr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8695) ["ID"]=> int(8695) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:39:56" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(3974) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 110JKR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_110jkr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_110JKR.pdf" ["wpmf_size"]=> string(6) "519169" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(107798) "Does Broadband Boost Local Economic Development? January 2010 Jed Kolko with research support from Davin Reed Supported with funding from The David A. Coulter Family Foundation http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 2 Summary “Broadband” refers to high -speed Internet service that —unlike dial-up modem service— is always “on.” This technology has become widely available throughout the United States. It is essential for businesses, and well over half of American households have broadband access at home. In recognition of its importance, public investment in broadband is surging. The American Recovery and Reinvestment Act (ARRA) of 2009 allocated $7.2 billion for broadband investment and commissioned a National Broadband Plan to promote universal access, foster economic development, and achieve additional potential benefits through thi s technology. Several California programs also support the expansion of broadband access, especially in areas where availability is lagging. This report assesses whether policies to raise broadband availability will contribute, as hoped, to local economic development. Our analysis relies on the fact that broadband technology has diffused unevenly throughout the United States, thus allowing us to compare economic indicators between areas with greater and less growth in broadband availability. Using broadban d data from the Federal Communications Commission and economic data from several government and proprietary sources, we examine broadband availability and economic activity throughout the nation between 1999 and 2006. Our analysis indicates a positive rel ationship between broadband expansion and economic growth. This relationship is stronger in industries that rely more on information technology and in areas with lower population densities. Although the evidence leans in the direction of a causal relations hip, the data and methods do not definitively indicate that broadband caused this economic growth. T he economic benefits to residents appear to be limited . Our analysis indicates that broadband expansion is also associated with population growth and that both the average wage and the employment rate —the share of working -age adults that is employed —are unaffected by broadband expansion. The economic benefits to households are thus more ambiguous than they would be if employment growth also led to an increas e in wages or the employment rate. We also found that expanding broadband availability does not change the prevalence of telecommuting or other home -based work. Of course, local employment growth might still raise property values and the local tax base , but in the absence of more direct benefits for residents in the form of higher wages or improved access to jobs, we can only say that the local economic development benefits of broadband are mixed. Broadband expansion may of course offer other social or economic benefits, such as improved health care delivery. Although our study does not examine such effects, we briefly review the limited evidence of other benefits in the conclusion of the report. Contents Summary 2 Tables 4 Figures 5 Introduction 6 Broadband Policies and Goals 8 The Federal Economic Stimulus and the National Broadband Plan 8 Broadband Policy in California 9 Broadband Availability and the Digital Divide 11 Unserved and Underserved Areas 11 Limitations of Available Broadband Data 14 What We Know about Broadband Availability 14 Why There Is a Digital Divide in Broadband Availability 15 Research Questions and Approach 17 Questions Addressed in This Report 17 Research Strategy 19 Questions This Report Does Not Address 21 Research Findings 22 Broadband and Local Overall Employment Growth 22 Broadband and Employment Growth across Industries and Places 23 Does Broadband Expansion Cause Employment Growth? 24 Broadband and Household Outcomes 25 Implications and Conclusion 28 What the Results Mean for Broadband as Local Economic Development Policy 28 Other Possible Benefits of Broadband 30 Broadband: The Next Generation 31 References 32 About the Author 34 Acknowledgements 34 All technical appendices to this paper are available on the PPIC website: http://www.ppic.org/content/pubs/other/110JKR_appendix.pdf http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 4 Tables Table 1. Broadband speed illustration 12 Table 2. Number of broadband providers by ZCTA 15 Table 3. Broadband and economic outcomes, 1999 –2006 22 Table 4. Broadband and industry employment growth, 1999 –2006 23 Table 5. Summary of findings 28 http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 5 Figures Figure 1. Broadband Availability in California, August 2009 13 http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 6 Introduction Since the mid -1990s, when businesses and households began to use the Internet, observers have been trying to predict —and more recently to assess —the effects of this relatively new mode of communication. Some of the early predictions about the Internet and related technologies, such as causing the “death of cities,” have clearly not materialize d. 1 The Internet has transformed many areas of life, providing individuals and businesses with easy and instant access to communication, information, and entertainment. But it may take years before the Internet —like many other new technologies —exerts a quantifiable effect on business productivity. It takes considerable time to develop applications and adjust business processes and organizations to take full advantage of new tec hnologies. But we still know very little about the economic effects of these technologies. 2 Nonetheless, governments need to assess whether new technologies are likely to offer benefits in the public interest and whether the private sector will make the technology available and affordable enough to support public policy goals. Occasionally, governments will deem a new technology essential for achieving these goals and will support universal access, even in areas that are expensive to serve. The federal government and the state of California, as well as other state and local governments , have made universal access to broadband service a public policy goal —the specific objective is to close the digital divide in broadband availability. 3 The phrase “digital divide” can refer to geographic inequalities in availability or to gaps in broadban d adoption owing to income, race/ethnicity, education, or other inequalities in access or skills that affect the ability of individuals or businesses to take advantage of broadband’s capabilities. 4 The federal and state broadband initiatives presume multiple economic and social benefits will accrue from increasing broadband access. Local economic development ranks high among these benefits . Other benefits include improvements in health care delivery, access to education, energy efficiency, civic participation, and public safety. To date, the evidence on the extent to which broadband provides any of these benefits has been quite limited. To help promote the goal of universal access, the federal government allocated $7.2 billion for broadband investment and also commissioned a National Broadband Plan as part of its 2009 American Recovery and Reinvestment Act (ARRA). These federal “stimulus” funds provided a large addition to existing state and lo cal efforts to support broadband access. In this report, we are particularly interested in the relationship between broadband and economic development. Previous research has not assessed whether broadband expansion causes economic development, nor has it examined who benefits from increased economic activity; there ha s been little research on the effect of broadband on outcomes other than employment, output, and income. Our analysis relies on the fact that broadband technology has diffused unevenly throughout the United States, thus enabling us to compare economic indi cators between areas with greater and less growth in broadband availability. In this report, we examine broadband availability throughout the nation between 1999 and 2006. Our parameters are partly determined by our 1 George Gilder, Forbes ASAP , February 27, 1995. Quoted in Mitchell Moss, “Technology and Cities,” Cityscape 3:3 . 2 “The Broadband Myth,” The Economist , May 23, 2008. 3 While broadband can include satellite and high -speed mobile data services, the vast majority of households and businesses that subscribe to broadband receive their service over cable, digital subscriber line (DSL), or other technologies that use physical wires or c ables to connect end- users to the Internet. Discussions about broadband availability usually re fer to wireline technologies. 4 “Digital divide” can refer to inequalities in the availability or adoption of other technologies as well. See Gunkel (2003). http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 7 data: Federal Communications Commission (FCC) data on broadband are available beginning in 1999, and our source for detailed employment data ends in 2006. We answer four questions: 1. Does employment grow faster in areas with greater broadband expansion? 2. Does the relationship between broadband and employment differ by industry or across places? For example, is it stronger for industries that are more reliant on technology or that use workers who are more technically knowledgeable? Is it stronger in places that are more isolated or in those with more amenities? 3. If there is a positive relationship between broadband and employment growth? Does broadband expansion cause employment growth? 4. If broadband does boost employment, who benefits? Is employment growth accompanied by a greater likelihood of employm ent, higher pay, increased income, or greater flexibility to be able to work from home? Although other studies have examined broadband and economic growth, this report offers more definitive answers by using richer data and more -refined methods. We also ex amine questions left unanswered in previous work, such as whether broadband actually causes economic growth, who benefits from this growth, and whether broadband increases telecommuting and other forms of home -based work. Of course, economic development is only one of many policy concerns that must be considered when targeting broadband investments, and in our concluding chapter we briefly review the limited evidence available on other potential benefits of this technology, including improved health care de livery and overall consumer welfare. We begin our discussion in the next chapter with a description of the national and California policies designed to increase broadband availability. We then describe the extent of broadband availability, explain why some places have better broadband access than others, and discuss the limitations of the available data. We then present our research questions, the methods we use to answer them, and the main findings of our study. We conclude with a discussion of the policy implications of our findings and then briefly review other possible effects of broadband that are outside the scope of our analysis. A technical appendix offers further details about related research and our data, methods, and results. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 8 Broadband Policies and Goals Broadband policy at both the national level and in California is designed to achieve multiple goals, most prominently economic development. The federal government, throug h Congress, the executive branch, and the Supreme Court, has shaped broadband policy through impo rtant regulatory decisions. 5 And today , federal public investment in broadband is surging. ARRA, the $787 billion economic stimulus package passed in 2009 , allocated $7.2 billion to broadband investments, with a particular focus on increasing broadband availability in targeted areas to improve economic development. 6 Well before the current federal investment in broadband, state and local governments have subsidized and in some cases directly provided broadband services. California has several programs in place that subsidize the demand and supply of broadband. The stated goals of these policies —described in detail below —include job creation and economic growth, civic participation, public saf ety, education, health care, and access to government services. Like federal policies, some California initiatives are more narrowly targeted toward underserved areas and economic development goals, while other programs are focused on a wider set of concer ns, such as health care and education. ARRA also requires the development of a National Broadband Plan to guide broadband policy beyond the federal stimulus funds. Th is p lan is supposed to be wide -ranging, covering numerous technologies, policy options, and goals. In sum, broadband policy at the federal level and in California aims to achieve a wide range of goals, with a particular emphasis —especially in the federal broadband stimulus —on raising broadband availability in unse rved and underserved communities in an effort to create jobs and stimulate economic growth. The Federal Economic Stimulus and the National Broadband Plan Signed into law in February 2009, ARRA directs two federal agencies, the Department of Agriculture’s Rural Utilities Service (RUS) and the Department of Commerce’s National Telecommunications and Infrastructure Administration (NTIA), to grant or lend $7.2 billion for broadband deployment and other broadband projects. 7 ARRA directs both agencies to conside r the effect on economic development when awarding funds. RUS will offer grants, loans, and loan guarantees worth $2.5 billion for broadband infrastructure projects in rural areas that lack “ sufficient access to high speed broadband service to facilitate rural economic development.” 8 5 The 1996 Telecommunications Act, the 2005 Supreme Court decision in National Cable & Telecommuni cations Association et al. v. Brand X Internet Services et al., and related FCC rulings helped create the national regulatory framework in effect today. Under this framework, telephone, wireless, television, and Internet providers are able to compete in an y market against each other, but broadband providers are not required to give competitors wholesale access to their infrastructures that would enable competitors to resell services to co nsumers. NTIA will award grants worth $4.7 billion to states, non- profits, and broadband providers for a wider set of broadband projects —not just infrastructure projects and not just in rural areas. While economic growth and job creati on are explicit goals of NTIA’s funds, as is raising broadband access in unserved and underserved areas, so is improved broadband access for schools, medical facilities, other community institutions, and 6 Of the $4 billion to be allocated in the first round of fun ding, 75 percent has been targeted toward infrastructure projects. 7 Applications for broadband stimulus funds are being accepted and reviewed in multiple rounds. First round applications were due in August 2009; up to $2.4 billion of the RUS funds and $1. 6 billion of NTIA funds will be awarded in this round. Subsequent rounds will award the remainder of the $7.2 billion by September 2010. Stimulus funds will be allocated by RUS through the Broadband Initiatives Program (BIP) and by NTIA through the Broadba nd Technology Opportunities Program (BTOP) . 8 ARRA, (H.R. 1) Title I, p. 4. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 9 public safety agencies. Together, these two components illustrate the prominence of economic development and job creation as goals for the broadband stimulus funds, among other goals. ARRA also requires the FCC, the national telecommunications regulator, to present a National Broadband Plan to Congress in F ebruary 2010 to guide future broadband policy. The P lan is supposed to “seek to ensure that all people of the United States have access to broadband ca pability” and is an acknowledgement that the broadband stimulus funds alone are “insufficient to support national broadband deployment.” 9 The Plan should recommend goals and benchmarks for a national broadband strategy and for achieving eleven enumerated policy goals, including job creation and economic growth, entrepreneurial activity, and other specific tar gets such as consumer welfare, civic participation, public safety, education, and health care delivery. 10 Broadband Policy in California The plan will thus cover a wider range of interests and technologies than the broadband stimulus funds, which are geared toward infrastructure, technol ogy mapping, and related activities. Well before the passage of ARRA, s tates and localities—including the state of California and some of its cities and counties —have made broadband investments through subsidies and direct p rovision. 11 As with federal broadband initiatives, California’s broadband policy is focused on numerous goals, with economic development prominent among them. 12 For many years, the state’s primary broadband subsidy program has been the Teleconnect Fund, esta blished in 1996, which pays half the cost of Internet access for qualified schools, libraries, community organizations, and other nonprofits, funded through a statewide fee on telephone service. The program has grown from $33 million in FY 2008– 2009 to $47 million (proposed) in FY 2009 –2010 and $67 million (proposed) in FY 2010– 2011. 13 Two newer programs have expanded the state’s involvement in broadband infrastructure deployment. In 2006, the California Public Utilities Commission (CPUC) created the Califo rnia Emerging Technology Fund (CETF), an independent nonprofit foundation. CETF has received funding for its first five years through a $60 million contribution from AT&T and Verizon, a condition of their respective mergers with telecommunications compani es SBC and MCI. CETF’s mission is to achieve “ubiquitous access to broadband and advanced services in California, particularly in underserved communities.” The foundation seeks to promote broadband availability in rural areas, affordability and adoption in urban poor and other disadvantaged areas, and accessible technology for people with disabilities. CETF hopes to disburse $240 million over five years, supplementing its own seed money with matching funds. As of December 2008, it had awarded $20 million i n grants. The largest award helped support the California Telehealth Network (CTN), which electronically connects hundreds of clinics and hospitals —primarily in rural areas and tribal lands —to medical centers. CETF’s support also helped CTN win a larger grant from the FCC’s Rural Health Care Pilot Program. 9 ARRA (H.R. 1), Section 6001 (k)(2), p. 402; FCC Notice of Inquiry 09- 31, 2009, paragraph 6. 10 The FCC’s Notice of Inquiry reviews these and other current FCC activities that relate to broadband. 11 Although the major current initiatives are sponsored at the federal and state levels, some localities in recent years have provided broadband directly t hrough public municipal Wi- Fi (wireless) networks, fiber-optic networks, or other broadband technologies. Kolko (2007) discusses local initiatives in California. 12 In addition to the subsidy and investment programs described in this section, California has set broadband policies through the final report of the California Broadband Task Force, created by Executive Order S -23- 06 in 2006, and the 2006 Digital Video and Infrastructure Competition Act (AB 2987), which created statewide video franchises and estab lished the CPUC’s responsibility for collecting and mapping broadband data. 13 Actual and proposed budgets are available at the CTF website, www.cpuc.ca.gov/PUC/Telco/Public+Programs/CTF/CTFList.htm . http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 10 The second new state program is the California Advanced Services Fund (CASF), which is providing $100 million over two years to subsidize broadband deployment in unserved and underserved areas; successful applicants are awarded 40 percent of total project costs for broadband infrastructure in these areas. CASF was authorized by the CPUC in 2007 and implemented in 2008 under SB 1193 (Padilla); the first projects were funded in November 2008. Whereas CETF ha s a broad strategy that includes infrastructure availability, adoption, applications, and technology literacy, CASF is more narrowly focused on promoting infrastructure availability in the largely rural parts of the state that lack adequate broadband. Thus , the main goal of CASF is quite similar to the objectives of the portion of the broadband stimulus funds targeted toward broadband availability in rural areas. The funds currently available through state programs —the Teleconnect Fund, CETF, and CASF—are small relative to the sums California might expect to receive under the federal broadband stimulus program. While the state programs have budgets or plans to disburse tens of millions of dollars annually, California’s share of the federal broadband funds w ould be over $900 million. 14 Thus, an essential element in California’s current broadband strategy is to win federal broadband stimulus funds. Numerous cities, counties, nonprofits, and companies in California were among the first -round applicants for federal funds: Their grant and l oan requests totaled $1.4 billion. 15 CPUC was a winner in the first round of the separate broadband mapping funds competition, receiving a $2.3 million grant in October 2009 . 16 California’s expectations, like those of the federal government, are that broadba nd will contribute to economic development. The final report of the California Broadband Task Force, for example, lists “economic and community development” first among the ways that broadband is likely to affect California, citing the potential effect of broadband on job creation in high- tech and other industries as well as on employment in rural areas. 17 The state also expects broadband to lead to increased telecommuting and environmental benefits. I n establishing its definition for “underserved” areas for CASF grant eligibility, the CPUC chose the minimum speed that would support telecommuting. 18 In turn, t he California Air Resources Board cites telecommuting as one way the state government can encourage its own employees to “decrease their individual carbo n impact.” 19 14 CETF’s “Summary of ARRA Proposals” estimates California’s “fair share” to be 13 percent of the federal total, proportional to the state’s share of the national population. 15 These applications have been submitted for the first round of $4 billion in federal funds, of which California’s 13 percent “fair share” would be $520 million —roughly one -third of the amount that California -based organizations have applied for. Applicants are listed at www.ntia.doc.gov/broadbandgrants/applications/search.cfm (viewed on October 12, 2009). The total request in California -based applications was reported in CETF’s “Summary of ARRA Proposals.” 16 The NTIA grant comes through the State Broadband Data and Development Grant Program, the program authorized by the 2008 Broadband Data Improvement Act and funded with up to $350 million from ARRA. 17 California Broadband Task Force, “The State of Connectivity,” 2008. Viewed on October 14, 2008, at www.calink.ca.gov/pdf/CBTF_FINAL_Report.pdf . Governor Schwarzenegger’s official response to the request for comments on the broadband stimulus also notes the importance of economic development: “ … with out any broadband service, [an area’s] residents are excluded from the economic development and social- economic benefits of broadband, including tele -education, telemedicine, access to health care information, access to government services and benefits, an d more.” See California response to BIP/BTOP request for comment, www.recovery.ca.gov/Content/Documents/California_Response_to_NTIA_on_ BB_Stimulus_Grants_4-13 -09.pdf , p. 12. 18CASF chose minimum speeds of 3 megabits per second downstream and 1 megabit per second upstream, arguing that at these speeds telecommuting becomes feasible, and lacking rapid enough connections to work from home qua lified an area as “underserved.” See Table 1 and California’s response to BIP/BTOP request for comment, www.recovery.ca.gov/Content/Doc uments/California_Response_to_NTIA_on_BB_Stimulus_Grants_4 -13 -09.pdf . 19 California Air Resources Board, Climate Change Draft Scoping Plan , June 2008, page 12. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 11 Broadband Availability and the Digital Divide California and federal policies seek to close the digital divide in broadband availability, placing a particular emphasis on unserved and underserved areas. ARRA and CASF define unserved and underserved in terms of whether broadband service is available and, if so, at what speed. CETF also mentions “ubiquitous access” and supporting “underserved communities” in its mission statement. Targeting “unserved” and “underserved” areas requires defining th ese terms and identifying which areas qualify. Recent improvements in data collection, and planned improvements funded by ARRA, will provide policymakers with much better information on broadband availability. However, for analyzing past trends in broadban d availability and the relationship between broadband and local economic development, we must rely on the less detailed data collected historically by the FCC. Unserved and Underserved Areas ”Unserved” and “underserved,” as defined for the federal stimulus funds, are an important policy statement about what constitutes adequate broadband availability. “Unserved” areas lack broadband service offering downstream speeds of at least 768 kilobits per second (kbps) and upstream speeds of 200 kbps. 20 These minimum speeds are lower than typical digital subscriber line (DSL) and cable broadband service today, which often advertise top downstream speeds of 1.5 to 6 megabits per second (mbps) and top upstream speeds of at least 768 kbps ( Table 1). “Underserved” areas ha ve either only partial service coverage at those speeds, no service at a higher speed threshold of 3 mbps downstream and 1 mbps upstream, or low adoption rates. 21 To assess the extent of the digital divide and to help identify unserved and underserved area s, the FCC recently improved its data collection efforts, and ARRA is allocating funds for additional mapping and surveying efforts. Since December 2008, broadband providers have been required to report subscribership levels by Census tract at different sp eeds of service through the FCC’s “Form 477” data collection process. (Form 477 data is about subscribership, and availability is inferred, based on the location of subscribers. This could result in an undercount of availability in areas where a provider has infrastructure allowing it to provide broadband service, but no subscribers.) As noted above, ARRA also seeks to provide information on unserved and underserved areas. It provides funding for the development of maps showing broadband availability and calls for the development of “a comprehensive nationwide inventory map of existing broadband service capability and availability” by February 2011. Finally, the 2008 Broadband Data Improvement Act requires the FCC to conduct consumer surveys on pricing, sp eed, adoption, and online behaviors, and to collect international data on broadband speeds and prices. 20 “Downstream” refers to data sent “down” from the Internet to an end -user, such as a downloaded file or a received email. “Upstream” refers to data sent “up” to the Internet from an end -user, such as an uploaded file or a sent email. 21 Unserved areas consist of contiguous Census blocks where at least 90 percent of households lack access to terrestria l (i.e., not satellite or mobile phone) broadband service offering speeds of at least 768 kbps downstream and 200 kbps upstream. Underserved areas consist of contiguous Census blocks where either (1) 50 percent of households or fewer have access to terrest rial broadband service of at least 768 kbps downstream and 200 kbps upstream; (2) no broadband service offers speeds of at least 3 mbps downstream and 1 mbps upstream; or (3) 40 pe rcent of households or fewer subscribe to broadband. Note that the standard of 3 mbps downstream / 1 mbps upstream is the level that CASF believes is necessary to support telecommuting, as mentioned in the previous section. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 12 TABLE 1 Broadband speed illustration Downstream speed Relevance Approximate time to download 3 megabyte pop song 56 kilobits/second (kbps) Top speed of dial -up modem service 7 minutes 200 kbps Minimum speed to qualify as “high-speed” (broadband) under historical FCC definition 2 minutes 768 kbps Minimum speed for an area not to be “unserved” according to ARRA 30 seconds 1.5 megabit s/second (mbps) Lo wer speed tier advertised for many broadband (DSL or cable) services 15 seconds 3 mbps Minimum speed for an area not to be “underserved” according to ARRA, and minimum speed CASF considers necessary for telecommuting 8 seconds 6 mbps Upper speed tier advertised for many broadband (DSL or cable) services 4 seconds 10 mbps Speed of advanced services available in some areas, including much of metropolitan southern California 2 seconds 100 mbps Speed of advanced services available in very limited areas, including parts of metropolitan Sacramento ¼ second NOTE: 1 byte equals 8 bits. Some states, including California, have already undertaken their own mapping initiatives. Based on infrastructure data collected from providers, Californi a publishes maps of broadband availability throughout the state, disaggregated into several speed tiers. 22 Maps dated August 10, 2009, show large rural and mountainous areas in the state without broadband access. The fastest service —more than 100 mbps— is a vailable in parts of the Sacramento metropolitan area; service at speeds of 10 –100 mbps is available throughout much of urban southern California, as well as Bakersfield and Napa and Solano Counties; and speeds of 5 –10 mbps are offered in most of the Bay A rea, including Silicon Valley (Figure 1). These differences demonstrate that even among places with broadband availability, speeds can vary considerably. And much of the state appears to have no service at speeds of at least 500 kbps, thus meeting the defi nition of unserved or underserved. 23 22 See maps at www.cpuc.ca.gov/PUC/Telco/Information+for+providing+service/Broadband+Availability+Maps.htm. 23 The speed categories presented in the CPUC map (Figure 1) do not accord with the speeds that define unserved and underserved areas (Table 1), making it difficult to determine which areas qualify for federal stimulus funds based on the map. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 13 FIGURE 1 Broadband A vailability in California, August 2009 SOURCE: California Public Utilities Commission, Communications Division, Video Franchising and Broadband Deployment Group. NOTE: Map used with permission of CPUC. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 14 Limitations of Available Broadband Data FCC data collected prior to the December 2008 improvements represent the best public historical information on broadband availability, and policymakers and academics have relied on these data to analyze the extent of broadband deployment and the relationship between broadband and economic and social outcomes. 24 These historical data consist of the number of subscribers at the state level, disaggregated by advertised speed of service and type of technology (cab le, DSL, etc.), as well as a list of ZIP codes where the provider has at least one subscriber to “high -speed Internet service,” defined as at least 200 kbps in at least one direction. Thus, the only sub -state information was the number of providers in a ZI P code with at least one subscriber, which has been published in semiannual reports back to 1999. 25 Some caution must be exercised in interpreting these data, because m any ZIP code s cover large geographic areas, and providers with a subscriber in a ZIP code might not offer service throughout the entire ZIP code. That said, these ZIP code provider -count data have two strengths for assessing policies designed to raise broadband availability. First, broadband policies are often enabled by adding providers to an area — sometimes directly through public provision and sometimes indirectly throu gh subsidization or regulation; thus, using the number of providers as a proxy for broadband availability is meaningful in a policy context. Second, there is a reliable relatio nship between the number of providers in a ZIP code and the estimated extent of residential broadband availability , even though the data are imperfect. As explained in the technical a ppendix, the publicly available FCC data do not distinguish between providers that own the wires and cables that constitute broadband’s infrastructure and those that lease and resell space on these lines to offer service; nor do FCC data indicate how much of the ZIP code each provider serves, nor at what speed. Nevertheless, related research has shown that estimated broadband availability in a ZIP code does increase when the number of providers increases, especially when the ZIP code has fewer providers to begin with. 26 What We Know about Broadband Availability Broadband is widely available. By December 2006, essentially all ZIP Code Tabulation Areas (ZCTAs) in the United States, including those in rural areas, had at least one provider offering service ( Table 2). 27 24 National broadband data collected prior to December 2008 cannot show which areas qualify as unserved or underserved, as defin ed for ARRA. Historically, the FCC has reported the number of providers in a ZIP code with service at speeds of at least 200 kbps in at least one direction, whereas an area qualifies as unserved if it lacks service of 768 kbps downstream and 200 kbps upstream. Thus, area s with “high-speed access” according to the historical FCC definition could still be unserved or underserved according to definitions for curren t broadband programs. This simple metric overstates broadband availability because providers do not always offer service throughout an entire ZIP code. Using an improved methodology that infers availability based on adoption patterns and FCC provider counts, we estimate that broadband was available to 85 percent of U.S. households and 92 percent of California households in December 2005 (Kolko, forthcoming). Looking at the number of providers, broadband availability continued to increase between 1999 and 2006, relative t o its level in 1999. The average number of providers per 25 These semiannual reports and provider count lists are available at www.fcc.gov/wcb/iatd/comp.html . Two reporting changes occurred starting in December 2005: First, providers with fewer than 250 high -speed connections were required to submit data through Form 477, whereas previously they were exempt; se cond, providers had to start reporting state -level subscribership by technology (GAO, 2009). 26 Kolko (forthcoming) demonstrates the relationship between provider count and availability. This relationship suggests that th e logarithm of the number of provide rs better approximates availability than either a linear measure or an assumption that availability reaches a maximum with one or a few providers. The appendix of this report discusse s other shortcomings with FCC Form 477 data in more detail, as does GAO (2006). 27 ZCTAs are U.S. Census Bureau approximations of U.S. Postal Service ZIP codes. ZCTAs are better suited to data analysis than ZIP codes. See the appendix for more detail about how ZCTAs and ZIP codes compare and how we converted ZIP code data to ZCTAs. “Rural” means outside a metropolitan area and accounts for roughly 20 percent of the U.S. population. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 15 ZIP code in the nation grew from 3.4 in 1999 to 11.2 in 2006, weighted by employment. The comparable figures for California were slightly higher (5.0) in 1 999 and similar (11.5) in 2006. TABLE 2 Number of broadband providers by ZCTA All areas Rural only U.S. California U.S. California 1999 Average broadband provider count 1.6 2.9* 0.8 0.9 Average broadband provider count (employment -weighted) 3.4 5.0* 1.7 1.6 Percent of ZCTAs with one or more providers 63% 77%* 42% 43% Percent of ZCTAs with one or more providers (employment - weighted) 96% 98%* 83% 81% 2006 Average broadband provider count 6.9 8.6* 5.0 4.9 Average broadband provider count (employment -weighted) 11.2 11.5 8.3 7.4* Percent of ZCTAs with one or more providers 99.6% 99.93%* 99.4% 99.5% Percent of ZCTAs with one or more providers (employment- weighted) 99.97% >99.99%* 99.94% 99.99% NOTES: Asterisks indicate California value is different from rest of U.S. at 5% statistical significance level. “Rural” means not in a metropolitan area. See technical appendix for further definitions and details. Since broadband appears, by these data, to be widespread, we may ask why federal and California broad- band policy is focused on unserved and underserved areas . The answer is that ARRA, CASF, and other current initiati ves use higher speed thresholds than the historical FCC definition of high- speed access (see Table 1 ). It is very likely that many areas are unserved or underserved relative to these thresholds, even if they have one or more broadband providers according t o the historical FCC definition. Thus, even though historical data on broadband availability do not show which areas qualify as unserved or underserved for the purposes of current broadband initiatives, they do reveal that broadband became available at di fferent times throughout the country. These historical data allow us to identify areas with more and less broadband availability and whether broadband expansion was associated with economic development outcomes. Why There Is a Digital Divide in Broadband Availability Why might broadband be more widely available in some places than others? The primary reason is that the costs and benefits of providing broadband depend on local factors. Broadband provision requires fixed costs to extend service to an area: Mu ch of the cost to install or upgrade telecommunications infrastructure is required “up front,” regardless of the number of eventual subscribers served by that infrastructure. Thus, in order to spread the fixed costs across more subscribers, providers are m ore likely to serve areas with high demand for broadband. In addition, infrastructure is more expensive to deploy in some areas, such as those with steep terrain or fewer roads, as broadband lines often follow existing transportation rights -of -way. Finally , broadband availability can vary because most areas in the United States are served by a dominant telephone provider and a dominant cable television provider, and each can make different strategic decisions http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 16 about when to introduce broadband service to the ir regions. State policies about regulating or subsidizing broadband could also affect the level of availability. 28 As expected, ZCTAs with higher population density, higher income, and flatter terrain had more broadband providers in 2006, holding other factors constant ( Technical Appendix Table A1 ). Roughly half of the difference between the number of providers in California and the nation in 1999 ( see Table 2 ) was due to differences in density, income, education, and terrain; the remainder was due to unmeasured factors, which could include state policies or the particular broadband strategies of the telephone and cable companies serving California. By 2006, the gap in availability bet ween California and the United States had closed to a statistically insignificant difference. The extent and reasons for the digital divide in Internet availability also depend on technology. Compared with today’s main broadband technologies, cable and DSL dial -up Internet service involves lower fixed costs and did not lead to persistent geographic disparities in availability. Fiber -to -the -home, which offers speeds much faster that DSL or cable, has high fixed costs of deployment and is therefore likely to lead to a more persistent digital divide than cable and DSL (Kolko, 2007). Even if today’s bro adband technologies become available everywhere, digital divides in future technologies will probably appear, and “closing the digital divide” will remain a policy goal. 28 We do not attempt to measure the effect of state policies on broadband availability because telecommunications regulations are difficult to quantify in a consistent way across states. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 17 Research Questions and Approach In this chapter we first discuss the questions this report addresses, explaining the underlying hypotheses and why the questions are important. We then present the empirical strategy that we use in answering these questions. We conclude the chapter with a discussion of other important questions about broad band’s effects that lie beyond both the scope of our analysis and the main concerns of current federal and state policy initiatives. Questions Addressed in This Report Our analysis answers four questions about broadband expansion and economic growth: 1. Does employment grow faster in areas with greater broadband expansion? 2. Does the relationship between broadband and employment differ by industry or across places? For instance, is it stronger for industries that are more reliant on technology or that use worke rs who are more technically knowledgeable? Is it stronger in places that are more isolated or that have higher amenities? 3. If there is a positive relationship between broadband expansion and employment growth, does broadband expansion cause this growth? 4. If broadband does boost employment growth, who benefits? Is employment growth accompanied by a greater likelihood of employment, higher income, or greater flexibility to be able to work from home? To answer our first question—whether the availability of broa dband is associated with local employment growth —it is helpful to consider why broadband might contribute to this type of local economic development. Obviously, broadband and other information and communications technologies lower the cost of sending and receiving many forms of data, including documents and audio and video content. According to standard economic theory, lowering the cost of one input to a profit -maximizing firm has two possible effects: First, the firm raises its output; second, the firm sh ifts its mix of inputs toward the input whose cost went down. When the spread of broadband lowers the cost of communication, the net result of these two effects on hiring is theoretically ambiguous. The first effect —raising output —would lead most businesse s to hire more labor. But the second effect —shifting toward the inputs that just got cheaper —could cause businesses to use new technology in lieu of labor for some tasks. Our second question explores how the relationship between broadband and employment v aries across industries, types of workers, and places. Again, it is useful to think about why there might be differences. Turning to industries first: Businesses reliant on information technology might increase their employment of workers skilled in using new technology and possibly even reduce their employment of others —shifting, in other words, toward labor that is “complementary” with broadband technology. Broadband would therefore have a larger positive effect on employment in industries whose workers are more skilled in using information technology. Furthermore, even if broadband expansion caused individual firms to reduce employment, economic activity could shift to locations where broadband is more widely available, raising aggregate employment there relative to other areas. Locational shifts in economic activity would be more pronounced in more “footloose” industries —those whose location is not tied to local markets or inputs. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 18 The relationship between broadband expansion and economic growth could als o vary across places. Broadband might offer greater benefit for places that are smaller or more isolated, helping local businesses or households to connect with larger markets: This line of thinking lies behind predictions that rural areas might benefit di sproportionately from Internet technology. Other examples include areas that have access to a more highly educated labor market (if the more educated workers are better able to use advanced information technologies) and areas with more favorable climates a nd recreational opportunities (if broadband access allows firms to move further away from suppliers and customers toward locations appealing to employers and workers). 29 At the same time, broadband expansion could lead to declines in both employment and e conomic output in local businesses that begin to face competition from online businesses located elsewhere. For example, broadband expansion might encourage households to purchase goods online instead of in local stores, or to download movies rather than visit their local theaters. As with the other possible links between broadband and local economic activity, the effects will vary by industry: Those businesses most dependent upon local demand, such as retailers or entertainment, are more likely to be hurt by broadband expansion, depending on how easily their customers can switch from local businesses to their online counterparts. Turning to our third question, it is important to acknowledge that a positive empirical relationship between broadband expansion and economic growth does not, in itself, mean that broadband expansion causes economic growth. The reverse might actually be true if broadband providers choose to offer or expand service in areas that are growing faster. Alternatively, population growth c ould cause both broadband expansion and employment growth. Broadband expansion might follow population growth since more than 60 percent of broadband subscribers are households, according to the most recent FCC broadband report. Once again, this effect cou ld vary by industry: Population growth would lead to employment growth in industries whose customers are local residents. Assessing causality is, of course, essential for predicting whether broadband policies will lead to economic development. Looking at b roadband expansion and employment growth in individual industries might help clarify the relationship between broadband expansion and overall employment growth. Our final question asks who benefits if, indeed, broadband expansion does cause economic growth. The answer depends, in part, on what happens to the population. If population growth accompanies employment growth because people follow jobs, then the likelihood of residents being employed— the employment rate— might not rise much. The effect on average pay and household incomes is unclear as well: Increased labor demand might raise wages, but if labor supply also increases, this would push wages down. Employment growth that raises the employment rate, average pay, or household incomes benefits residents; employment growth that does not might still contribute to property owners’ land values or local governments’ tax base, benefitting some residents but not others. 30 It is quite plausible that targeting unserved and underserved areas for broadband expansion could raise local employment yet offer ambiguous economic returns for residents. Economists are often skeptical about “place -based” policies, which provide geographically targeted infrastructure investments; the broadband 29 “Why Wall Street Is Losing Out To 40 Acres and a Modem,” New York Times, December 27, 1998, highlights the relocation of some financial operations to idyllic recreation areas like Jackson Hole and Nantucket. 30 We do not assess the relationship between broadband expansion and property values because the time period we study (19 99–2006) coincides with the large and varying house price bubbles in many local housing markets. The geographic variation in price changes durin g this time is probably too noisy to be a reliable measure of the geographic variation in the capitalization int o land values of potential local productivity enhancements such as broadband expansion. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 19 stimulus funds are an example of s uch policies. According to Louis Winnick (1966), an economist and urban development expert, “Federal programs to change the geography of output are a kind of welfare device to redistribute personal income. But at best it is a clumsy, expensive, and often inequitable device. Not only are the gains to one locality offset by losses to another, but even in the locality of gain the added income frequently goes to the wrong people. … A disproportionately large share of the increased purchasing power goes to the owners of immobile resources [e.g. property owners] other than labor.” 31 Furthermore, place - based policies may end up encouraging economic activity in places where that activity might not otherwise be economically sustainable: Many of the places that are unserved and underserved by broadband are so because terrain, remoteness, or low population density raises the cost of broadband provision. 32 Research Strategy Yet, in defense of place- based policies, not all people can or want to leave disadvantaged places, and governments are committed to offering basic services to residents in all places. The relationship between broadband and economic outcomes at the local level is relevant for assessing policy that targets specific locations for broadband investments. As described above, broadband availability has varied across the nation, and this report takes advantage of these geographic differences to assess how the expansion of local broadband availability relates to changes in many economic outcomes. Several other studies have also looked at the relationship between broadband and economic development. They have generally found a positive relationship between broadband expansion and employment growth and a mixed relationship between broadband expansion and income (or wage) growth. By using richer data, better measurement, and a wider range of methods, this report offers more definitive answers about the relationship between broadband and economic growth, which are generally consistent with previous work. 33 Our research strategy consists of comparing changes in several economic outcomes wi th the extent of broadband expansion. To measure the expansion of broadband availability, we use the FCC’s Form 477 data on the number of broadband providers in a ZIP code, as described in the previous section and the This report also examines questions left unanswered in previous work, such as causality, the effect on population, and the effect on telecommuting and other forms of home -based work. technical appendix . We match these data with economic outcomes from several other data sources:  The change in employment provided in the National Establishment Time -Series (NETS) database, which reports employment for nea rly all businesses in the United States from 1992 to 2006 and includes detailed industry and geographic information. The NETS is based on the Dun & Bradstreet business register.  The change in employed residents, total and working -age population, average pa y, and median household income, all at the county level, from the U.S. Census and other government data sources.  The change in the likelihood of telecommuting, bringing work home, and operating a home -based business, as reported in surveys conducted by For rester Research, a technology consultancy. Forrester surveys households annually about technology adoption and related behaviors. 31 Ladd (1994), in a review of research on enterprise zones, also notes that the benefits of place -based policies to non -landowning local residents are uncertain. 32 One prominent urban economist (Edward Glaeser) argues that disaster- struck New Orleans (Glaeser, 2005) and economically depressed Buffalo (Glaeser, 2007) should be allowed to shrink, with support given directly to residents rather than to places. 33 The appendix offers more detail on related academic and policy research. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 20 The FCC, NETS, and Forrester all report information at the ZIP code level. ZIP code areas are small relative to counties and offer rich detail on local broadband availability and economic outcomes. But official ZIP codes, as defined by the U.S. Postal Service (USPS), change frequently over time and do not accord precisely with county boundaries, Census tracts, or other areas for which data are typically available. Furthermore, while the NETS and Forrester report actual USPS ZIP codes for businesses and households, respectively, the FCC uses its own approximation of ZIP codes. And the Census uses its own ZIP code approximation, the ZCTA (ZIP code tabulation area), for reporting selected demographic and economic measures. We converted data from all these sources to the Census ZCTA definitions to create consistent geographic boundaries across datasets and over time. Most previous research has examined the relationship between broadband and employment growth at larger geographic levels (e.g., counties or states). However, using smaller units of geography is more desirable because broadband availability can vary block -by -block. ZIP codes are the smallest geographic area for which both broadband and employment data are publicly available. 34 We focus on broadband expansion and outcomes between 1999 and 2006. Although broadband diffusion began earlier, most growth in the number of providers occurred during this period (the FCC began reporting provider count data in 1999). These years are also most relevant as a guide to what future broadband expansion—through ARRA or other programs —might mean for economic development. The relationship between broadband (or any technology) and economic outcomes might change over the course of the technology’s diffusion. 35 Policies designed to bring broadband to still -underserved areas are at the end of the broadband diffusion process (at current broadband speeds , anyway), so more recent experience is a better guide than earlier experience to what might happen in the future. 36 The first step, then, is to assess the overall relationship between broadband expansion and employment growth. We then look at whether the relationship varies by industry or type of place. Next, we try to assess ca usality by examining the industry -specific effects and by also using an “instrumental variable” strategy. To further refine these insights, our analysis highlights the relationship over the period 1999 –2006 in ZCTAs with the least broadband provi sion in 1999. Although data do not yet exist to assess whether the locations that lagged in 1999 are the same locations that remain unserved or underserved by today’s policy definitions, this analysis may offer guidance for the effects of policies targeting today’s unserved and underserved areas. 37 34 To assess the relationship between broadband expansion and employment growth, the ideal level of geography would be th e exact street address because a business can have broadband access only if it is available at its exact location. The ZCTA is the smallest geography for which data are available and is therefore the level of analysis we use for employment growth. In contrast, counties are more appropriate than ZCTAs for examining household labor market outcomes because counties better approximate the size of labor markets and broadband could affect households that are in the same labor market as businesses that adopt broad band. And finally, we examine other outcomes: population growth, changes in the likelihood of being employed, average pay, and household income, as well as the changing prevalence of three types of home -based work: telecommuting, bringing work home occasionally, and operating a home -based business. These additional outcomes reveal how broadband expansion relates to outcomes that households experience. 35 It is ambiguous whether the economic effects, theoretically, might be larger earlier or later in the diffusion process. Early users, by taking advantage of a technology first, might grow to a scale that later users would find difficult to compete with. However, the cost of adopting a new technology can fall over time with technological improvements and knowledge from lessons learned by earlier users, which migh t increase the economic benefits for later users. 36 Focusing on the 1999 –2006 period also allows us to adjust for earlier trends in employment growth, because the NETS data start in 1992. Adjusting for prior trends is important because it accounts for the possibility that earlier employment growth encouraged lat er broadband expansion, and late r employment growth might simply be the continuation of the earlier growth trend rather than the effect of broadband expansion. 37 The instrumental variable strategy identifies a factor – in this case, slope of terrain – that affects broadband expansion wi thout independently affecting employment growth, holding other factors constant. The relationship between employment growth and the variation in broadband expansion that is predicted by slope identifies the causal portion of the effect of broadband on grow th, at least for the areas in which slope is a good predictor of broadband expansion. The appendix offers detail on this approach and the results. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 21 Althou gh our analysis focuses on the 1999 –2006 period, there is also some value in looking at the 1992 –1999 period, when broadband first became available in some areas. 38 Questions This Report Does Not Address The relationship between broadband expansion and economic growth in this earlier period may offer guidance on how policies aimed at hastening the diffusion of early -stage technologies affect local economic development, assuming that next -generation technologies that fundamentally change business processes will have the same relationship to econom ic activity as did current broadband technology. Because federal and state broadband policies are particularly focused on unserved and underserved areas, this report looks at how local broadband expansion relates to l ocal economic outcomes. The methods used in this report and in related research examine the relationship between broadband expansion and local activity relative to other areas. Even if broadband had no net effect on employment growth nationally, we could s till see a positive relationship between broadband expansion and local employment growth if broadband expansion shifted economic activity from some locations to others, resulting in no net gains in jobs or output. If equity is an important public policy go al, governments could still favor broadband investments that redistribute economic activity toward underserved areas. Yet broadband policies could also affect economic activity nationally. Numerous countries have set goals not only to close the digital div ide by making a minimum broadband service available to all residents, but also to support higher -speed broadband services to keep high -tech industries globally competitive (Li and Losey, 2009). In its own evaluation of broadband deployment in the United St ates, the FCC considers both the extent and evenness of broadband access among Americans —the digital divide—as well as comparisons with other countries, though there is disagreement about which measures accurately reflect international differences in broadband availability and, accordingly, disagreement about the international ranking of the United States. 39 Broadband may also offer benefits not ful ly captured in measures of output, employment, or growth. People use broadband for a wide range of activities, including many —such as sharing pictures with friends or downloading music—that might not fulfill a public policy goal, even though people value t hese activities. Other benefits —for example, access to news, remote medical services, or distance learning —might be public policy goals but might not be reflected in standard economic indicators such as employment, output, or income. Thus, looking at the r elationship between broadband and economic outcomes might exclude some benefits of broadband that do achieve public policy goals and others that do not. Although our analysis in this report focuses on local economic development measures, we also review the limited recent research on other outcomes in our final chapter. It is thus challenging to estimate the net national benefits of broadband, and attempting to do so requires different methods. Although this subject lies beyond the scope of our report, we do return to the question of national benefits briefly in the conclusion, as measured by consumer willingness to pay for broadband. 38 As the appendix explains, a methodological disadvantage of looking at the 1992 –1999 period is that we have to assume that no ZIP code had any broadband providers in 1992, so the level of provi ders that the FCC reported in 1999 equals the change between 1992 and 1999. Another disadvantage of looking at the 1992 –1999 period is that we cannot adjust for prior trends in employment because the NETS data start in 1992. 39 FCC 08 -88, Fifth Report on br oadband deployment, as required by Section 706 of the 1996 Telecommunications Act, June 2008. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 22 Research Findings Broadband and Local Overall Employment Growth The overall relationship between broadband expansion and employment growth, as measured by the NETS, is positive. Moving from no broadband providers to 1 –3 providers (the FCC groups one, two, and three providers together in its reporting) is associated with employment growth that is higher by 6.4 percentage points over the seven -year period from 1999 to 2006 (Table 3). 40 The size of this relationship is large relative to the overall national employment growth rate, but employment growth at the ZCTA level shows wide variation, and broadband expansion accounts for relatively little of the variation in growth rates across ZCTAs. 41 technical appendix tables The relationship between broadband expansion and employment growth is similar when looking only at areas with few or no broadband providers in 1999. As explained above, these less -well -served ZCTAs in 1999 are the best guide to the relationship between future broadband expansion and employment growth in the areas targeted today by federal and state programs to raise broadband availability. Broadband expansion is associated with a smaller but still statistically significant increase in employment growth of 3.5 percentage points in the 1992 –1999 period. (Many of the results discussed in this section are shown in the .) TABLE 3 Broadb and and economic outcomes , 1999–2006 Percentage point change associated with increase in broadband availability Employment growth (ZCTA) 6.4* Working age population (county) 2.4* Employed residents / working age population rate (county) -1.2 Average pay per employee (county) 1.1 Median household income (county) -2.4* NOTES: Percentage changes reflect the coefficient of the change in the dependent variable, in lo g form, for a shift from 0 to 1– 3 providers or from 1– 3 to 4 providers, or equivalent changes in the log of the number of providers. Asterisks denote significance at the 5% level. Employment growth is from the NETS; remaining variables are from Census and BLS. See technical appendix for detailed explanation and complet e results. 40 Our regression analysis uses the change in broadband providers in logarithmic form, so adding more providers has a less -than -proportional additional effect on employment and other outcomes. See appendix for details. 41 The cumulative national employment growth rate in 1999 –2006, according to the NETS, was 0.1 percent, which is well below Census estimates based on business surveys (6.2%) and household surveys (7.5%). The NETS is known to undercount some employment in the most recent years, though local employment growth across counties is highly correlated between the NETS and Census business surveys, especially when looking over multiyear employment changes rather than one -year changes. See Kolko and Neumark (2007) for d iscussion of the NETS and its comparison to other data sources. The standardized beta of the relationship is .08, meaning that a one standard -deviation increase in broadband availability is associated with a .08 standard deviation increase in employment growth. See appendix for details. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 23 Broadband and Employment Growth a cross Industries and Places The relationship between broadband expansion and employment growth varies across industries. The positive relationship is especially large for utilities; information; finance and insurance; professional, scientific, and technical services; management of companies and enterprises; and administrative and business support services. The relationship in these sectors is much larger than the relationship for overall employment (Table 4) and is positive and statistically significant for all but two sectors: mining and public administration. TABLE 4 Broadband and industry employment growth, 1999–2006 Employment growth Percentage point employment change associated with increase in broadband availability Highest share of technology inputs Highest share in computer occupations Employment location most tied to population Management of companies and enterprises (55) 40.8*** X X Utilities (22) 16.7*** X Professional, scientific, and technical services (54) 16.4*** X X Finance and insurance (52) 14.8*** X Administrative and business support services (56) 14.1*** X Information (51) 12.0*** X X Construction (23) 11.8*** X Agricultural, forestry, fishing, and hunting (11) 11.6*** Real estate and rental and leasing (53) 10.2*** X Accommodation and food services (72) 9.9*** Transportation and warehousing (48-49) 8.6*** Health care and social assistance (62) 7.4*** Wholesale trade (42) 7.1*** Other services (81) 7.1*** X Mining (21) 6.6 Retail trade (44-45) 6.5*** X Manufacturing (31-33) 6.3** Educational services (61) 6.1*** X X Arts, entertainment, and recreation (71) 5.7** Public administration (92) 0.5 NOTES: Numbers in parentheses are the NAICS codes for the industry sector. Industries are ranked by the employment change associated with in creased broadband availability. Percentage changes reflect the coefficient of the change in the dependent variable, in log form, for a shift from 0 to 1 –3 providers or from 1 –3 to 4 providers, or equivalent changes in the log of the number of providers. Asterisks denote significance at the 5% level. Employ ment growth is from the NETS. See technical appendix for detailed explanation and complete results. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 24 The relationship between broadband and employment tends to be stronger in industries where information technology (IT) services (Internet publishing, telecommunications services, data processing, and related services) represent a larger share of an industry’s inputs. These industries include: information; professional, scientific, and te chnical services; management; administrative services; and educational services. Of these, all but educational services are among the industries whose employment growth shows the strongest relationship with broadband expansion. In addition, industries wit h a larger share of employees in computer specialist occupations tend to show a stronger relationship between broadband expansion and employment growth. Utilities; information; finance and insurance; professional, scientific, and technical services; and ma nagement had a higher share of employees in these occupations, and all five of these industries are among those showing the strongest relationship between broadband expansion and employment growth. 42 A possibility raised above is that some businesses could be hurt by broadband expansion if online services compete with traditional businesses. Retail is one example, if people with broadband access choose to shop online instead of at local retailers. Another could be the arts, entertainment, and recreation industry, whic h could see less local demand for live or on -site events if broadband makes online substitutes available. Both of these industries show weaker relationships between broadband expansion and employment growth, relative to other industries. In sum, industries that rely more on technology inputs an d on workers in computer specialist occupations— the industries that should benefit more from broadband —are those in which broadband expansion is associated with stronger employment growth. The relationship between broadband and employment growth is also stronger in some places than others. For example, the relationship is stronger for ZCTAs with lower population density (and, conversely, weaker for those with higher density) —consistent with the theory that s maller or more isolated areas may benefit more from high -speed connections, giving businesses in these areas access to larger markets. However, even for most high density areas, the relationship between broadband and growth remains positive on balance, jus t not as large as for lower -density areas. None of the other place characteristics —such as having a more educated workforce, having a better climate, or being a vacation destination —affects the relationship between broadband expansion and employment growth ( Technical Appendix Table A4 ). Does Broadband Expansion Cause Employment Growth? A crucial question for broadband policy is whether broadband expansion causes employment growth: If so, then policies to make broadband available should result in local growth. Conceivably, a positive relationship between broadband expansion and employment growth could arise for other reasons. For example, if broadband providers ex pand in locations where they anticipate future growth, then the positive relationship would in part or entirely reflect this strategic decision of providers rather than a causal effect of broadband on growth. Alternatively, population growth could cause both broadband expansion and employment growth: Broadband providers could invest in areas where population (and therefore demand for broadband) is growing, while at the same time population growth could cause employment growth in industries (such as retail, restaurants, and personal services) that serve local populations. 42 The appendix explains these two industry -level measures of technology intens ity. “Inputs” are based on the Bureau of Economic Analysis’s input -output tables, which describe the types of inputs – including IT services –each industry uses in producing its output. Occupational data, by industry, are from the Bureau of Labor Statistics. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 25 Our results seem to rule out the idea that broadband providers are explicitly targeting areas where they expect higher economic growth: Later employment growth does not predict earlier broad band growth. The evidence also indicates that population growth is not the main driver. Adjusting the analysis for population growth changes the relationship between broadband expansion and employment growth only minimally: The boost to employment growth f alls from 6.4 to 5.0 percentage points but remains statistically significant ( Technical Appendix Table A2 ). Furthermore, the sectors whose employment is most tied to population are not those that show the strongest relationship between broadband expansion and employment growth. 43 A more technical assessment of causality uses the instrumental variable approach described above, in which slope of terrain in a ZCTA is used as prox y for broadband expansion that should not have any independent relationship with employment growth. Jobs tend to locate where people do in industries such as construction, real estate rental and leasing, education, retail trade, and other personal services; t hese industries are not among those in which employment grows most where broadband expands ( see Table 3). 44 Technical Appendix Table A5 Using this approach, the relationship between broadband expansion and employment growth remains positive and statistically significant, suggesting a causal relationship. However, using this approach, the size of the relationship appears implausibly large and is quite sensitive to how the regression model is specified, which makes us less confident in this result ( ). This approach, therefore, is only modestly suggestive that broadband expansion causes employment growth. These different approaches, though not definitive, generally point in the direction of a caus al relationship and suggest that broadband expansion leads to large increase in local employment growth. Broadband and Household Outcomes To assess how the employment growth associated with broadband expansion benefits households, we consider key labor market outcomes. First, we look at the employment rate: the percentage of working -age residents that are employed. Other important outcomes are average pay per employee and median household income, which could change with employment growth because of changes in labor demand, labor supply, or the composition of the workforce. 45 Broadband expansion has no statistically significant relationship with the employment rate. As discussed above, it does not appear that population growth causes both broadband expansion and employment growth. Rather, people may be following jobs: The employment growth associated with broadband expansion could encourage people to move to areas where employment opportunities are expanding. Furthermore, although most people work in their county of residence, some people commute across county lines. In counties where many employees live elsewhere or where many residents work elsewhere, broadband expansion is asso ciated with an increase in employment but not in the number of employed Finally, we consider whether broadband availability facilitates working from home. 43 The appendix explains this industry -level measure of whether employment is tied to population, which is based on the similarity between the geographic distribution of employment in an industry and the geographic distribution of population. 44 As the appendix explains, the regression includes variables such as road density that, if omitted, could lead slope and employment growth to have a direct relationship and would invalidate this approach. 45 The analyses of household outcomes look at the county, rather than the ZCTA, level. Although businesses should be directly af fected by broadband expansion only at their e xact address, increased demand for employment could benefit workers throughout a labor market, and counties —which are almost always larger than ZCTAs —are a better approximation of labor markets than ZCTAs. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 26 residents. In other words, some new jobs associated with broadband expansion aren’t filled by local residents: Employment opportunities attract people who are willing to move or commut e. 46 Broadband expansion is associated with no change in average pay per employee and a decrease in median household income. The results for counties with less broadband availability in 1999 —the group which may be most comparable to the places that current broadband policy targets—are similar: There is no relationship between broadband expansion and either the employment rate or average pay per employee. In sum, whatever positive effects broadband may have on employment growth, it did not result in either h igher employment rates or higher pay for residents in areas where broadband expanded during the period 1999 – 2006 period. As described above, these residents may benefit indirectly from employment growth in their county if that growth raises the local tax b ase or property values, but they do not benefit directly in terms of greater likelihood of being employed or higher earnings or income. Residents who rent, and therefore might have to pay more for housing when property values rise, could possibly become worse off economically from employment growth that raises neither the employment rate nor average pay per employee. The household outcomes reveal a difference between the earlier and later time periods of broadband’s diffusion. In contrast to the later period, broadband expansion from 1992 to 1999 showed positive and statistically significant relationships with average pay per employee, median household income, and the employment rate ( Technical Appendix Table A6 ). One possible explanation is that businesses that adopted broadband earlier faced a labor market where computer skills were less widespread than in later years; these early adopters might also have had to hire workers with more advanced skills if they had to develop more applications in -house to integrate a nascent and fundamentally new technology into their business processes. Later adopters could rely more on off -the -shelf mass -market applications that workers with more modest technology skills could use. 47 Finally, we examine the link between expanded broadband availability and the likelihood of working at home. At -home work could benefit households by giving them (and their employers, in the case of telecommuting) more flexibility. We find no relationship between broadband expansion and all three types of home -based work: (i) having a formal relationship with an employer to work at home at least one day a week (which the survey refers to explicitly as “telecommuting”); (ii) bringing work home to do outside of normal work hours; and (iii) operating a business from home ( If the supply of workers with the appropriate technology skills were more limited in the earlier period, that could result in both higher increases in the employment rate and in average pay per employee in areas with greater b roadband expansion. Technical Appendix Table A7 ). Even for the types of people whose jobs or skills might lend themselves more easily to home-based work —people with college degrees or in managerial or professional occupations —there was no relationship between broadband expansion and changes in doing home -based work. The absence of a relationship between broadband expansion and home -based work may seem surprising, yet recent experience shows that telecommuting —even in places where technology makes telecommuting possible —is rarer than some have expected. One reason for this might be that communication with remote work ers through “telepresence” or other video communications often involves expensive hardware and 46 Note that the number of employed residents and the employment rate (reported by the Census) are based on whether residents of a county are employed, regardless of where their employer is located. In contrast, employment growth (reported by the NETS) is based on wh ere businesses are located, regardless of where their employees live. See appendix for further discussion. 47 The NETS data provide information on the number of employees in a business but no information about their education or skill level. Data on the characteristics of workers in firms adopting broadband technology at different stages of broadband diffusion would help a ssess these conjectural explanations. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 27 much faster broadband connections than the minimum speeds that qualify as “broadband.” Recall that the speed necessary for telecommuting according to the CPUC —3 mbps downstream—is fifteen times faster than what qualifies as broadband in the historical FCC definition. Another plausible explanation is that corporate culture can deter telecommuting, even if all of the requisite technology is available: Telecommuters often feel invisible to their employers, and managers often feel a loss of control and uncertainty about the output of remote workers. 48 The relationship between broadband and telecommuting could change, however, with the development of low -cost, high- quality video technology and management styles more conducive to remote work. 48 “Home Warriors: Telecommuters Need More Than E- mail and a Broadband Connection,” The Economist , July 25, 2008. “Top 10 Myths About Mobile Work,” Sun Microsystems Open Work Services Group White Paper, 2008. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 28 Implications and Conclusion What the Results Mean for Broadband as Local Economic Development Policy The findings indicate a positive relationship between broadband expansion and em ployment growth (Table 5). The evidence suggests that broadband may have a causal effect on employment growth: The relationship is strongest in more technology -reliant industries, and population growth does not appear to be the trigger, which was the most plausible alternative explanation to a causal relationship because most broadband subscribers are households, not businesses. TABLE 5 Summary of f indings Economic outcome Relationship with broadband Possible reason Employment Positive and big Broadband expansion causes existing businesses to expand or redistributes economic activity toward the area Working-age population Positive People are mobile and move to where employment opportunities are expanding Employment rate (employed residents / working-age population) None Because people are mobile or willing to commute, l abor supply grows along with labor demand, and businesses need not pay a large premium for basic technology skills Average pay per employee None Median household income Negative Telecommuting None Home-based work requires faster speeds than the minimum that qualifies as broadband, and many corporate cultures are not well-suited to telecommuting Bringing work home None Having home-based business None NOTE: Employment data are from NETS; population, employed residents, employment rate, average pay, and median household income data are from Census/BLS; telecommuting, bringing work home, and having home- based business data are from Forrester. However, the large increase in employm ent growth associated with broadband expansion does not necessarily benefit local residents. Areas with faster broadband expansion between 1999 and 2006 experienced no greater increases in either the employment rate—employment as a share of the working -age population —or in average pay per employee, relative to other areas, and median household income declined. One possible explanation for this is that broadband does indeed lead to employment growth, which encourages people to move or commute to areas where employment opportunities have expanded, and this increase in the local labor supply prevents the increased demand for labor from raising either the employment rate or average pay. Employment growth might still raise local property values and tax bases, but in the absence of more direct benefits for residents, the economic development benefits of broadband are ambiguous. As discussed above, place -based policies —for example, targeted broadband infrastructure investments —often involve uncertain economic benefi ts for residents. This debate surrounding place -based policies such as broadband infrastructure investment, along with our finding that broadband expansion raises neither the employment rate nor the average pay of residents, begs the question of whether pu blic money designated for broadband infrastructure might have a larger effect on http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 29 economic or social outcomes if the funds were allocated instead toward subsidizing broadband adoption or other needs of disadvantaged households, regardless of where they live . There is no easy answer; but it depends to some extent on the contribution of other potential benefits of broadband, which we discuss below, and how much society values investment in disadvantaged places. One caution with regard to our findings is that the best publicly available data on broadband’s diffusion — the FCC’s count of providers in a ZIP code —is an imperfect measure of availability. Although the provider count measure we use in our study is a good proxy for availability and an improvement over previous studies, it does not take broadband speed into account, which can vary considerably across different locations. This limitation may be especially relevant for our finding of no association between broadband expansion and any of three types of home -based work, including telecommuting. While broadband service that meets the minimum speed in FCC’s historical definition of broadband may not raise the prevalence of telecommuting, future broadband services t hat support videoconferencing and other “telepresence” applications might do so. Better broadband data will make the relationship between broadband speed and economic outcomes clearer in future research. Improving Data for Measuring and Evaluating Broadband The federal stimulus includes significant funds for broadband mapping and data collection, and the National Broadband Plan has the potential to propose additional forms of broadband research. We recommend two research directions to further the understanding of broadband and its effects. First, we should not only measure the availability, adoption, and use of broadband but also identify potential benefits of broadband and assess whether broadband diffusion is meet ing those goals. ARRA is focused on measuring access to broadband but not its effects. Second, although the FCC has already begun to collect better data on broadband availability since December 2008, earlier data on broadband availability —including the data we use in this report —could be much improved with better geographic specificity and information about speeds. To the extent that the FCC or other agencies have collected these data in the past but have not made them publicly available, they should do so. With a better understanding of where exactly and at what speeds broadband became available, researchers can more accurately assess the relationship between broadband and economic development and other outcomes —and offer better guidance to policymakers abo ut how to maximize the benefits of future broadband policies. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 30 Other Possible Benefits of Broadband Although economic development is a primary goal of ARRA and other initiatives seeking to raise broadband availability, ARRA, in the National Broadband Plan, directs the FCC to consider other possible benefits of broadband, including consumer welfare, health care, education, civic participation, government services, energy independence and eff iciency, and public safety. 49 Very few studies have examined the effect of broadband on these other outcomes. 50 The research on consumer welfare tries to place a dollar value on broadband access, based on what consumers are willing to pay. Yet estimates vary wildly beca use of different assumptions about how many residents value their home broadband connections well above the $30 –50 per month that many households pay, and how much these people would actually be willing to pay. The estimates of the consumer benefit of broa dband range from a few billion dollars per year to hundreds of billions per year, with little way to assess which assumptions about consumer willingness to pay are more accurate. The most compelling research in these areas either estimates the consumer welfare of broadband based on demonstrated willingness to pay for servi ce or considers how switching from dial -up to broadband Internet access changes online behaviors. 51 Recent research shows that people switching from dial-up to broadband spend more time using the Internet but that only some online behaviors change (Kolko, 2009). Yet even if it were possible to pin this number down, people might value bro adband for activities without any clear public benefit, such as downloading music, rather than for activities often believed to be in the public interest, such as increasing civic participation. To answer the policy question of why government should suppor t broadband, the fact that consumers want broadband is not enough; consumers value many goods and services, such as cars and cable television, which governments do not subsidize. The case for government support of broadband depends on what consumers use broadband for and what effects broadband has. 52 Other research has identified improved health care as a realistic benefit of broadband. One study estimates that broadband will save $30 billion per year in reduced medical costs and reduced institutionalization due to electronic medical records, remote monitoring of health indicators, and other health applications that depend on broadband (Litan, 2005). Policies have already begun to reflect these possible benefits of Switching to broadband increases the likelihood and intensity of several activities that do not fall under goals promoted by broadband policy: downloading music, purchasing products online, and visiting adult entertainment sites. Among “socially desirable” online activities that do fall under these goal s, the only activity that increases is researching health information. There is no change in visits to job or career websites or government websites, including civic participation activities such as getting information about public hearings or contacting e lected officials. Changing online behaviors will not necessarily lead to better social outcomes. For example, researching health information online will not necessarily enable people to live longer or healthier lives. But it is difficult to see how broadba nd will come to affect policy goals such as civic participation without changing online behaviors. 49The FCC’s Notice of Inquiry asks whether broadband could contribute to energy independence and efficiency through teleworking, a “smart” electricity grid, or “intelligent” highways that monitor traffic. Fuhr and Pociask (2007) suggest numerous ways that broadban d and other information technologies could achieve environmental and energy goals, including better supply -chain management, fewer printed materials, and reductions in driving and business travel. 50 Some research has looked at the relationship between Internet access in general and outcomes such as community participation and social ties, racial discrimination in retail transactions, book prices, and other outcomes, but not at broadband specifically. Since dial- up access has long been available throughout the United States, the relevant question for policy is the effect of having broadband access relative to ha ving dial- up access. Studies looking at the effect of Internet access generally do not help us understand the marginal effect of broadband. 51 See the appendix for more detail on willingness -to -pay estimates by Crandall, Jackson, and Singer (2003) and Greenstein and McDevitt (2009). 52 Other studies, including Anderson (2008) and Hitt and Tambe (2007), also found that broadband adopters increase their time on line, but the researchers’ me thodology did not allow them to identify effects on specific online activities. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 31 broadband on health. CETF’s largest award to date has been to the California Telehealth Network (CTN), connecting clinics, hospitals, and medical centers throughout the state. At the federal level, the FCC’s Rural Health Care Pilot Program began in 2007 to allocate over $400 million for broadband infrastructure and design for health care purposes; this program also supports CTN. Although we point to health care as a potential area of benefit, the evidence so far is based primarily on online behavi ors. Further studies are needed to assess the relationship between broadband availability and social outcomes, even if these effects are too long -term to be immediately measurable: health outcomes could include illness incidence or mortality rates; educati on outcomes could include attainment of a high school diploma or college degree. If future studies show evidence of broadband’s benefit for health or other social outcomes, broadband policies could consider weighing multiple criteria in determining which g eographic areas might benefit the most from closing the digital divide. For instance, policies could explicitly take into account not only which areas have the least broadband availability, but also which areas have the most pressing need for improvements in health care or another outcome that future research discovers is improved by increasing access to broadband. Broadband: The Next Generation Federal and state broadband policies seek to invest in broadband in underserved areas to close the digital divide. Yet closing the digital divide is not the only broadband policy that could affect economic development outcomes; and policies that support and hasten early -stage rollout of next -generation technologies, such as extremely fast fiber- to-the -home or new te chnologies that affect business processes, could have different economic development outcomes than those we report. What does appear certain is that even if the United States achieves the policy goal of ubiquitous broadband availability at current speeds, and the current broadband digital divide closes, new digital divides will open. Fiber -to -the -home, for example, is likely to promote even more extreme geographic disparity in availability than current broadband because the fixed costs of provision are so high (Kolko, 2007). Recent maps of broadband availability in California already show disparities across the state in available speeds. As average speeds increase, new applications develop that take advantage of —and require—these higher speeds, making what may have been an adequate broadband connection in the past inadequate for applications deemed in the future to be important for participating in the economy or society. Federal and state governments might again consider policies to increase the availabilit y of new technologies in less-well -served areas, expecting specific economic or social benefits from the expansion of the new technology. Our research on recent broadband expansion shows that technologies that contribute to local employment growth might no t, in fact, benefit local residents; and we must carefully consider who would ultimately benefit from public investments in next -generation broadband and other new technologies. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 32 References Anderson, Ben. 2008. “The Social Impact of Broadband Household In ternet Access.” Information, Communication, and Society 11 (1): 5 –24. California Public Utilities Commission . 2006. ConnectingCalifornia: Broadband Report Update. Available at ftp://ftp.cpuc.ca.gov/PUC/Telco/Reports/california+broadband+report+for+sept+2006+cetf+meeting.pdf Crandall, Robert, Charles Jackson, and Hal Singer. 2003. “The Effects of Ubiquitous Broadband Adoption on Investment, Jobs, and the U.S. Economy.” Criterion Economics, L.L.C. Released by New Millennium Research Council. Available at http://newmillenniumresearch.org/archive/bbstudyreport_091703.pdf. Crandall, Robert, William Lehr, and Robert Litan. 2007. “The Effects of Broadband Deployment on Output and Employment: A Cross -sectional Analysis of U.S. Data. ” Issues in Economic Policy discussion paper no. 6. Brookings Institution. Available at http://www.brookings.edu/~/media/Files/rc/papers/2007/06labor_crandall/06labor_crandall.pdf. Federal Communications Commission . 2006. “High -Speed Services for Internet Access: Status as of December 31, 2005 .” Available at http://fjallfoss.fcc.gov/edocs_public/attachmatch/DOC -266596A1.pdf. Flamm, Kenneth . 2009. “Diagnosing t he Disconnected: Where and Why I s Broadband Access Unavailable in the U.S.?” Unpublished manuscript . Forman, Chris, Avi Goldfarb, and Shane Greenstein. 2009. “The Internet and Local Wages: Convergence or Divergence?” NBER Working Paper #14750 . Fuhr, Joseph, and Ste phen Pociask. 2007. “Broadband Services: Economic and Environmental Benefits.” American Consumer Institute. Gillett, Sharon, and William Lehr. 1999. “Availability of Broadband Internet Access: Empirical Evidenc e.” Unpublished manuscript . Gillett, Sharon, W illiam Lehr, Carlos Osorio, and Marvin Sirbu. 2006. “Measuring the Economic Impact of Broadband Deployment.” U.S. Department of Commerce, Economic Development Administration . Glaeser, Edward. 2005. “ Should the Government Rebuild New Orleans, or J ust Give Residents Checks?” The Economists’ Voice 2 (4): article 4. Glaeser, E dward. 2007. “Can Buffalo Ever Come Back? ” City Journal 17 (4). Available at http://www.city - journal.org/html/17_4_buffalo_ny.html . Glaeser, Edward, Hedi Kallal, Jose Scheinkman, and Andrei Shleifer . 1992. “Growth in Cities.” Journal o f Political Economy 100 (6). Government Accountability Office. 2006. Broadband Deployment Is Exten sive throughout the United States, but It Is Difficult to Assess the Extent of Deployment Gaps in Rural Areas. GAO -06-426. Report to congressional committees . Government Accountability Office. 2009. Current Broadband Measures Have Limitations, and New Measures Are Promising but Need Improvement. GAO -10-49. Report to congressional committees. Greenstein, Shane, and Ryan McDevitt. 2009. “The Broadband Bonus: Accounting for Broadband Internet’s Impact on U.S. GDP.” NBER Working Paper 14758 . Grubesic, Tony . 2003. “Inequit ies in the Broadband Revolution. ” Annals of Regional Science 37:263–89. Grubesic, Tony. 2006. “A Spatial Taxonomy of Broadband Regions in the United States.” Information Economics and Policy 18. Grubesic, Tony, and Alan Murray. 2004. “Waiting fo r Broadband: Local Competition and the Spatial Distribution of Advanced Telecommunication Services in the United States.” Growth and Change 35 (2). Gunkel, David. 2003. “Second Thoughts: Toward a Critique of the Digital Divide .” New Media & Society 5 (4): 499–522. Henderson, Vernon, Ari Kuncoro, and Matt Turner. 1995. “Industrial Development in Cities. ” Journal of Political Economy 103 ( 5) . Hitt, Loren, and Prasanna Tambe. 2007. “Broadband Adoption and Content Consumption.” Information Economics and Policy 19 (3–4): 362–78. http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 33 Kolko, Jed. 2000. “The D eath of Cities? The Death of Distance? Evidence from the Geography of Commercial Internet Usage.” In The Internet Upheaval: Raising Questions, Seeking Answers in Communications Policy , e d. Ingo Vogelsang and Benja min M. Compaine (Cambridge, MA: MIT Press), 73 –98. Kolko, Jed . 2007. “Broadband for All? Gaps in California’s Broadband Adoption and Availability .” California Economic Policy 3 (2). Available at http://www.ppic.org/content/pubs/cep/EP_707JKEP.pdf . Kolko, Jed, In press. “How Broadband Changes Online and Offline Behaviors.” Information Economics and Policy. Kolko, Jed. Forthcoming . “A New Measure of U.S. Residential Broadband Availability.” Telecommunications Policy . Kolko, Jed, and David Neumark. 2007. Business Location Decisions and Employment Dynamics in California. San Francisco: Public Poli cy Institute of California . Ladd, H elen. 1994. “Spatially Targeted Economic Development Strategies: Do They Work?” Cityscape: A Journal of Policy Development and Research 1 (1): 193−218. Li, Chiehyu, and James Losey. 2009. “100 Megabits or Bust! An Overview of Successful National Broadband Goals from Around the Globe.” Policy paper, New Americ a Foundation. Available at http://www.newamerica.net/publications/policy/100_megabits_or_bust . Litan, Robert. 2005. “Great Expectations: Potential Economic Benefits to the Nat ion from Accelerated Broadband Deployment to Older Americans and Americans with Disabilities. ” New Millennium Research Council . Available at http://newmillenniumresearch.org/arc hive/Litan_FINAL_120805.pdf. Osorio, Carlos. 2006. “The Economi c Impact of Municipal Broadband.” Unpublished manuscript . Prieger, James. 2003. “The Supply Side of the Digital Divide: Is There Equal Availability in the Bro adband Internet Access Market?” Economic Inquiry 41 (2). Shideler, David, Narine Badasyan, and Laura Taylor . 2007. “The Economic Impact of B roadband Deployment in Kentucky. ” Regional Economic Development 3 (2): 88–118. Song, Moohoun, Peter Orazem, and Rajesh Singh. “Broadband Access, Telecommuting, and the Urban -Rural Digital Divide. ” Iowa State Department of Economics Working Paper #06002. Stenberg, Peter, Mitchell Morehart, Stephen Vogel, John Cromartie, Vince Breneman, and Dennis Brown. 2009. “Broadband Internet’s Value for Rural Am erica.” U.S. Department of Agriculture, Economic Research Service. Van Gaasbeck, Kristin, Stephen Perez, Ryan Sharp, Helen Schaubmayer, Angela Owens, and Lindsay Cox. 2007. Economic Effects of Increased Broadband Usage in California. Summary report, Sacram ento Regional Research Institute. Wallsten, Scott. 2009. “Understanding International Broadband Comparisons: 2009 Update.” Technology Policy Institute. Winnick, Louis. 1966. “Place Prosperity vs. People Prosperity: Welfare Considerations in the Geographic Redistribution of Economic Activity.” In Essays in Urban Land Economics , by James Gillies et al. (Los Angeles: University of California Real Estate Research Program). http://www.ppic.org/main/home.asp Does Broadband Boost Local Economic Development? 34 About the Author Jed Kolko is an associate director of research at the Public Policy Institute of California, responsible for managing the institute's economy research. He has conducted numerous studies of the California economy, economic development, housing, and technology policy. Prior to coming to PPIC in 2006, he was vice president an d research director at Forrester Research, a technology consultancy, where he managed the company’s consumer market research businesses and served as the lead researcher on consumer devices and access technologies. Jed has also worked at the Office of Fede ral Housing Enterprise Oversight, the World Bank, and the Progressive Policy Institute. He holds a Ph.D. in Economics from Harvard University. Acknowledgements I thank Gary Bjork, Michael Byrne, Ellen Hanak, William Lehr, David Neumark, Anne Neville, Micha el Teitz, and Lynette Ubois for thoughtful reviews, suggestions, and insights. I am grateful to Davin Reed for excellent research support. Participants at the 2008 Telecommunications Policy Research Conference offered helpful suggestions on an earlier vers ion, and Forrester Research graciously gave me permission to use its Technographics survey data. PUBLIC POLICY INSTITUTE OF CALIFORNIA Board of Directors Walter B. Hewlett , Chair Director Center for Computer Assisted Research in the Humanities Mark Baldassare President and CEO Public Policy Institute of California Ruben Barrales President and CEO San Diego Chamber of Commerce John E. Bryson Retired Chairman and CEO Edison International Gary K. Hart Former State Senator and Secretary of Education State of California Robert M. Hertzberg Partner Mayer Brown LLP Donna Lucas Chief Executive Officer Lucas Pub lic Affairs David Mas Masumoto Author and farmer Steven A. Merksamer Senior Partner Nielsen, Merksamer, Parrinello, Mueller & Naylor, LLP Constance L. Rice Co- Director The Advancement Project Thomas C. Sutton Retired Chairman and CEO Pacific Life Insura nce Company Carol Whiteside President Emeritus Great Valley Center The Public Policy Institute of California is dedicated to informing and improving public policy in California through independent, objective, nonpartisan research on major economic, social, and political issues. The institute’s goal is to raise public awareness and to give elected representatives and other decisionmakers a more informed basis for developing policies and programs. The institute’s research focuses on the underlying forces shaping California’s future, cutting across a wide range of public poli cy concerns, including economic development, education, environment and resources, governance, population, public finance, and social and health policy. PPIC is a private operating foundation. It does not take or support positions on any ballot measures or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. PPIC was established in 1994 with an endowment from William R. Hewlett. Mark Baldassare is President and Chief Executive Officer of PPIC. Walter B. Hewlett is Chair of the Board of Directors. Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright n otice is included. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Copyright © 2010 Public Policy In stitute of California All rights reserved. San Francisco, CA PUBLIC POLICY INSTITUTE OF CALIFORNIA 500 Washington Street, Suite 600 San Francisco, California 94111 phone: 415.291.4400 fax: 415.291.4401 www.ppic.org PPIC SACRAMENTO CENT ER Senator Office Building 1121 L Street, Suite 801 Sacramento, California 95814 phone: 916.440.1120 fax: 916.440.1121" ["post_date_gmt"]=> string(19) "2017-05-20 09:39:56" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(8) "r_110jkr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:39:56" ["post_modified_gmt"]=> string(19) "2017-05-20 09:39:56" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(50) "http://148.62.4.17/wp-content/uploads/R_110JKR.pdf" ["menu_order"]=> int(0) ["post_mime_type"]=> string(15) "application/pdf" ["comment_count"]=> string(1) "0" ["filter"]=> string(3) "raw" ["status"]=> string(7) "inherit" ["attachment_authors"]=> bool(false) }