Leveling the Internet Playing Field

EXECUTIVE SUMMARY

Many public libraries and school districts are slow to adopt Internet connections, while some are more rapid. This difference is leading to an inequity between students and library patrons in different districts, leading to a competitive disadvantage for those students, at least in the short run and possibly in the long run. This inequity could be overcome by means of state government action in the form of cost-sharing. This report estimates the costs that would be involved.

This report relies on several important assumptions and findings from other reports in this series. In particular:

• Dial-in Internet access by modem is widely available in Kansas. However, modem access provides an inferior level of service. In particular, its high cost leads to very limited hours of access per client. It also restricts the types of use that ca n be made of the Internet. Direct connections to the Internet provide much better service than modem connections, but only a few districts have provided them as yet.

• Inequality could be analyzed from the taxpayer's point of view, as differences in the cost per client of providing access. Instead, this report takes the client's point of view, and analyzes inequality as differences in the level of access actual ly provided. This report defines "equal access to the Internet" as the existence in each Kansas district of direct connections to the Internet with at least one connected computer per 30 students plus one per teacher in each school building, and one compu ter per librarian plus one per 2000 citizens served in each library. Because Internet access in a time of rapidly changing technology is a moving target, the report assumes that all equipment is replaced after five years, so that equipment is 2.5 years be hind the cutting edge, on average.

• Only 21 Kansas school districts with direct connections to the Internet were identified as of February 1996. These districts were disproportionately large and metropolitan, but the number of these districts is so small that a detailed statistical analysis would not be meaningful.

• At the same time, a meaningful statistical analysis can be performed on "advanced" school districts. We define "advanced" school districts as those with high speed area networks, and/or with direct connections to the Internet, either in existence or planned for 1996. Providing an area network is an important precursor to providing direct connections to the Internet. Many schools districts adopt area networks first for reasons independent of Internet access.

• "Advanced" school districts are disproportionately large, metropolitan, densely populated, and wealthy, both in terms of assessed value per pupil and budget per pupil. However, a detailed, statistical analysis shows that the most important influe nces on network adoption are population density and assessed value per pupil.

• Behavioral models of local school board budget decisions were developed using "median voter" assumptions and Kansas data. Models were developed both for being "advanced" and for providing a given total school budget per pupil.

• The models show that a 10% increase in tax costs leads to a 5% reduction in school expenditures per pupil, while a 10% increase in assessed value per pupil leads to a 3% increase in school expenditures per pupil.

• Differences across locations in access costs per client are important causes of differences in the level of access actually provided. However, local wealth and local attitudes are also important.

• These models show that achieving near-universal access to the Internet will depend on taking steps toward:

  1. Equalizing costs per pupil of Internet access.
  2. Helping districts with low wealth and income.
  3. Helping laggard districts to get started.

• Exactly equalizing per pupil costs and per patron costs across regions would require around $22 million annually. However, it may be neither necessary nor even desirable to equalize per-client costs exactly. A practical degree of cost-equalizatio n can probably be achieved for $4 million per year.

• Some $10-15 million per year, at most, would be required to overcome differences in wealth and overcome attitudes of resistance to providing Internet access. (That is in addition to an assumed $4 million in cost-equalization.) In order to keep ab reast with rapidly changing Internet technology, these expenditures would need to continue over time.

Return to The List of Research Reports