A r c h i v e d  I n f o r m a t i o n

Investing in School Technology: Strategies to Meet the Funding Challenge - November 1997

Chapter 2 (continued)

The Economics of Accessing the Information Superhighway

In this section, we explore the underlying cost structures of the models for accessing the information superhighway examined earlier. A basic understanding of the economics of these models is necessary prior to applying the cost analysis framework presented in the section that follows. The data in this section rely on cost estimates produced by Rothstein and McKnight 13 and are used to illustrate some basic differences in the economics of these models.

Exhibit 2.5 shows that, excluding Model 1, which provides minimal benefits, there is relatively little difference in the annual operating (or ongoing) costs of Models 2, 3, and 4. Even though annual operating costs are fairly similar, on an a priori basis it seems clear that Model 4 (the Classroom model) should generate substantially greater benefits than Models 2 and 3 (the Laboratory-based models). This is because Model 4 offers better connectivity to the information superhighway and greater student and teacher accessibility to computers. In addition, the installed network associated with Model 4 is able to support a wider range of applications. Model 5 also offers potential benefits, but the start-up costs are significantly higher than in other models. The relative differences in start-up (or one-time) costs between Models 2, 3, and 4 are larger than is the case with ongoing costs. For example, Exhibit 2.5 shows that the start-up cost of Model 4 is almost four times higher than Model 2, and over twice as high as Model 3. However, is possible to design a funding strategy that defrays or spreads Model 4's high initial costs over a period of years. Although the start-up costs of Model 4 are four times higher than those for Model 3, Exhibit 2.5 shows that on an annualized basis Model 4 is less than twice as expensive (only 90 percent more expensive) as Model 3.

This brief analysis of the economics of different models of accessing the information superhighway contains an important lesson for the funding strategy: as the capacity (bandwidth) and size (number of nodes) of the network increases, initial or start-up costs increase much more rapidly than annual operating costs. This is because of the higher costs associated with increased computer penetration and the much larger technical support and staff training required. In other words, more bandwidth and nodes mean greater start-up costs but still manageable maintenance costs. The funding strategy should focus on keeping these start-up costs to a minimum while maintaining the level of service and support required by the technology plan.

Undisplayed Graphic

The funding strategy may also need to focus on overcoming the potential barrier to implementation imposed by high initial costs. In these circumstances, the technology plan should be designed so that it gives the school district the flexibility to spread start-up costs over several years. This type of strategy may include phasing in different schools at different times; phasing in and evaluating implementation of different models at a school over time (i.e., moving from one stage of technology to the next); leasing equipment rather than purchasing it outright; and folding start-up costs into bond issues. Some school districts are now emulating practices found in parts of the corporate sector. In the airline industry, for example, the heavy training costs associated with introducing a new aircraft type are typically financed as part of the aircraft acquisition package and amortized over several years in the balance sheet.

Appraising the Technology Plan

In the final analysis, a school technology plan will be evaluated based on student results. But it also is important for school leaders to appraise the technology plan in terms of what strategies could fund it. The cost analysis framework presented below allows us to do this. By helping to portray the cost dynamics of the technology plan, the framework can be used to assess whether the demands placed on funding can be assuaged either by reconfiguring the main elements of the plan or by reconsidering the timing of its implementation. The latter point is important because school districts should not ignore the fact that two of the most powerful variables in the funding equation are directly under their control, i.e., the individual elements that make up the technology plan and the timing of the plan's implementation. Analyzing technology costs using this type of framework will greatly facilitate subsequent development of a school district funding strategy.14

The Cost Analysis Framework

As stated earlier, the purpose of developing a cost analysis framework is to obtain detailed estimates of the costs of the proposed technology plan over time in order to assess funding requirements. The costs that are initially developed represent the baseline case: the technology plan designed to achieve the school district's education goals. If the results of the analysis reveal there is pressure on funding, it may become necessary to compare the costs of possible alternative configurations of the plan, including an alternative implementation schedule. A comprehensive cost analysis framework also can be used to determine the impact on project cost (and funding requirements) of alternative financing options or cost assumptions. These detailed estimates of the costs of developing and operating the plan will provide the designers or managers of the school district's technology system with the information necessary to formulate alternative approaches to funding the plan. Therefore, instead of simply trying to attach a price tag to the plan, the purpose of the cost analysis framework outlined in this section is to help determine viable ways to fund the plan.

An accurate cost profile of the school district's technology plan can be captured by ensuring that the cost framework distinguishes each individual cost item in the plan by: (a) school; (b) technology component; and (c) the year in which each cost will be incurred. The annual cost of each item in the plan will be estimated for each year in the system's projected life cycle. The timing of expenditures, and the amount, should be captured in terms of when actual cash outlays take place. Our proposed cost analysis framework is illustrated, in general form, in Exhibit 2.6.

Cost Analysis Framework Exhibit

The framework distinguishes the three stages in a system's life cycle: development (i.e., planning and design), implementation, and operation. In the case of a school district technology project, the planning and design task normally will be completed several months before implementation of the system (i.e., Yrt-1, where Yr1 represents the first year of the plan's operation). All the one-time, non-recurring costs that are expended before the system becomes operational, such as staff training, installation of local or wide area networks, or equipment purchases, are allocated by convention to time period Yr0 (day one of the system's first year of operation). The sum of all expenditures occurring in Yr0 and Yrt-1 represents the project's start-up costs. Recurring or ongoing costs are incurred throughout the system's life cycle. They are allocated to the year in which they are expected to be incurred (Yrn). Clearly, recurring costs predominate in the operations stage of the system's life cycle. Some representative cost categories, organized by technology component, are included in Exhibit 2.7. These costs should be projected at the school level, and later aggregated to form a profile of costs at the school district level.

In projecting future costs, the school district should use constant dollars, even though current dollars are normally used in budget projections. Constant dollars are current dollars that have been adjusted for the effect of inflation on prices. Unlike for most production goods, the prices (in current dollars) of many technology items actually fall over time. When converted from current to constant dollars (i.e., after discounting for the effects of inflation), the price of these items will be even lower. 15 On the other hand, the price of some technology items will remain constant or increase due to the addition of new capabilities that buyers deem important to acquire.

When comparisons are being made of the costs of alternative strategies, the future streams of costs (in constant dollars) must also be discounted to their "present value." 16 Present value calculations discount future uncertainty and thereby equalize the comparison of alternative investments when expenditures or revenues are distributed unequally over time. The current year establishes the time reference point for present value calculations. The present value calculation discounts or reduces the costs projected to occur in future years to a common point in time (i.e., the present) so they can be compared on a common basis. Present value analysis is based on the principle that costs that occur in the future are less burdensome than the same level of costs that occur now, and that money promised in the future is worth less than the same amount of money obtained now.

After the annual cost of each cost element at the school level has been projected, the system's life-cycle cost profile can be prepared. Exhibit 2.8 provides an example of the type of worksheet that can be used to develop this profile. To develop the school district cost profile for the baseline case, one of these worksheets should be prepared for each school in the district and for the district office. The costs contained in each worksheet are then aggregated to produce the cost profile for the whole school district. Charges for services shared among the district’s schools can be prorated. Spreadsheet software, such as Lotus, Quattro, and Excel, provides an ideal tool for this type of analysis. A simplified summary of a school district's cost profile, based on the framework illustrated in Exhibit 2.6, is shown in Exhibit 2.7.

Exhibit 2.7A
School District Cost Profile
(Baseline case, current $000s)
(i.e. priced at the year of expenditure)

Plan component

Yr-1

Yr0

Yr1

Yr2

Yr3

Yr4

Yr5

Total

Network & file servers

291

4420

844

870

896

923

950


Computers, other hardware, sw

0

6000

257

265

273

282

290


Technical support & maintenance

0

0

1545

1591

1639

1689

1738


Staff development

243

1500

360

371

383

394

406


Facility modification & wiring

194

1250

0

0

0

0

0


Yearly total cost

728

13170

3006

3097

3191

3288

3384

29864

Total system cost in current dollars = 29.864 million

Exhibit 2.7B
School District Cost Profile
(Baseline case, constant $000s)
(i.e. compensating for inflation of 3%)

Plan component

Yr-1

Yr0

Yr1

Yr2

Yr3

Yr4

Yr5

Total

Network & file servers

300

4420

820

820

820

820

820


Computers, other hardware, sw

0

6000

250

250

250

250

250


Technical support & maintenance

0

0

1500

1500

1500

1500

1500


Staff development

250

1500

350

350

350

350

350


Facility modification & wiring

200

1250

0

0

0

0

0


Yearly total cost

750

13170

2920

2920

2920

2920

2920

28500

Inflation adjusted value = $28.500 million 17in constant dollars.

Exhibit 2.7C
School District Cost Profile
7 Year Present Value of Total Cost Based on Constant Dollars and a 4% Discount Rate

Yearly total cost

750

13170

2920

2920

2920

2920

2920

28500

Present Value

781

13170

2808

2670

2596

2496

2400

26921

Present value = $26.921 million

In projecting future costs, the school district should use constant dollars, even though current dollars are normally used in budget projections. Constant dollars are current dollars that have been adjusted for the effect of inflation on prices. Unlike for most production goods, the prices (in current dollars) of many technology items actually fall over time. When converted from current to constant dollars (i.e., after discounting for the effects of inflation), the price of these items will be even lower.18 On the other hand, the price of some technology items will remain constant or increase due to the addition of new capabilities that buyers deem important to acquire.

When comparisons are being made of the costs of alternative strategies, the future streams of costs (in constant dollars) must be discounted to their "present value."19 Present value calculations equalize the comparison of alternative investments when expenditures or revenues are distributed unequally over time. The current year establishes the time reference point for present value calculations. The present value calculation discounts or reduces the costs projected to occur in future years to a common point in time (i.e., the present) so they can be compared on a common basis. An equivalent calculation is also made to bring the value of past expenditures (e.g. Yr-1) to today’s values. Present value analysis is based on the principle that costs that occur in the future are less burdensome than the same level of costs that occur now, and that money promised in the future is worth less than the same amount of money obtained now. 20The discount rate to be used is generally the "cost of money." For the purpose of the examples noted in Exhibit 2.7C and 2.9, this is assumed to be 4 percent over the rate of inflation. If the present value calculation were to be run against the "current value" then a 7 percent discount rate would be used. When run against the "constant value," a 4 percent discount figure is used, as inflation has already been taken into account.

Exhibit 2.8
Systems Life Cycle Profile

__Baseline or __Alternative _____ Year __Constant Dollars or __ Current Dollars
Cost Category Year -1 Year 0 Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 System Life Total
Non-Recurring Costs








  Equipment Purchase
  & Fees









  Installation








  Software Purchase








  Personnel








  Training








Subtotal








Recurring Costs








  Equipment Lease
  and Maintenance









  Software Lease
  and Maintenance









  Personnel Salaries/
  Benefits









  Direct Support Services








  Training








  Supplies








  Utilities








  Security (Incl. Backup)








  Overhaul








Subtotal








TOTAL Projected Costs








TOTAL Present
Value Costs









CUMULATIVE Total
Project Costs









After the annual cost of each cost element at the school level has been projected, the system's life-cycle cost profile can be prepared. Exhibit 2.8 provides an example of the type of worksheet that can be used to develop this profile. To develop the school district cost profile for the baseline case, one of these worksheets should be prepared for each school in the district and for the district office. The costs contained in each worksheet are then aggregated to produce the cost profile for the whole school district. Charges for services shared among the district’s schools can be prorated. Spreadsheet software, such as Lotus, Quattro, and Excel, provides an ideal tool for this type of analysis. A simplified summary of a school district’s cost profile, based on the framework illustrated in Exhibit 2.6, is shown in Exhibit 2.7.

Exhibit 2.7C shows that the present value of the cost of the district’s technology plan over seven years (Yr-1 through Yr5) is $26.9 million. However, the profile of costs shows that most costs (almost $14 million) are incurred in the first two years of the life cycle, even before the system becomes fully operational. The system’s ongoing costs are relatively modest. On the basis of these results, if it appeared the district would have difficulty funding the plan’s large start-up costs, refinements or modifications to the plan could be modeled using the same cost profile framework. Any alternative cost profiles produced in these circumstances should be assessed not only in terms of whether they assuage the initial funding requirements but also in terms of whether, on the basis of present value, they result in higher overall project costs. Exhibit 2.9 illustrates this type of comparison.

In Exhibit 2.9 the usable "life" used for the calculation is seven years, the first year being for planning. In all of the options it is assumed that replacement and upgrade will continue for the foreseeable future. Future years with zero cost is therefore not anticipated for the purpose of this comparison.

Exhibit 2.9
Comparing the Annual and Total Costs of Different Cases
(Constant $ Millions - 7 year life - 4% Discount)


Yr-1

YrO

Yrl

Yr2

Yr3

Yr4

Yr5

PV-$M

Baseline

.75

13.17

2.92

2.92

2.92

2.92

2.92

$24.92

Case 1

.75

10.07

4.47

4.47

2.92

2.92

2.92

$24.75

Case 2

.75

3.07

5.99

5.99

5.99

5.99

2.92

$25.88

Case 3

.75

8.80

2.35

2.35

2.35

2.35

2.35

$18.53

Case 1: Consists of implementing the technology in the district’s schools at different times over three years; 2 buildings and district office at Yr0, one building in Yr1 and one building in Yr2 (total =$19.01M, including recurring cost) which is the same as the Baseline case for the first 3 years

Case 2: Assumes that the network, computer equipment, and associated items are leased rather than purchased; lease terms are 60 months at 6.25%; an additional $2.92 M/Yr is added for recurring expenses.

Case 3: Amends the technology plan to provide a lower computer penetration ratio.

If the large up-front cash outlays in the baseline case present an insuperable funding problem, the cost of alternative configurations of the technology plan can be modeled and evaluated using the cost analysis framework. Three examples of alternative configurations of the baseline technology plan are shown in Exhibit 2.9. In Case 1, the technology is implemented in different schools at different times. It assumes that the district office and two schools implement the plan first (in Yr0), followed by one other school in each of the succeeding two years. The initial cost for the plan is the same as if all the work were done in the first year (in constant dollars), but this strategy reduces outlays in the first year of the project by spreading the cost of implementation over three years. On the basis of overall present value, compared to the baseline case, this strategy produces a lower overall project cost but delays the availability of program benefits to all students.

Case 2 assumes the school district arranges for lease financing for the listed components of the technology program. This strategy smoothes the cost of implementation over the life of the project and results in the lowest initial cash outlay over the first five years compared to the baseline and Case 1 strategies, and is only slightly more expensive when measured in life time cost (on the basis of present value). This option allows all the students to participate in the technology program from its inception while providing the flexibility to change out and upgrade equipment.

Case 3 is based on a more drastic action, involving a reduction in the degree of computer penetration (e.g., by installing fewer computers per classroom or by installing the same number of computers per classroom as the baseline case but reducing the number of classrooms in which they are installed). This strategy will produce lower equivalent costs ($18.53 million overall), but it also reduces the project’s expected benefits and may impair the plan’s ability to facilitate the district’s educational goals. In any event, if the principal concern about funding is the plan’s high initial cash outlay, then case 2 provides a better option.

Preparing for the Funding Plan

A successful funding plan will attack the funding challenge from all possible perspectives so as not to place undue reliance on any one strategy. Planning funding strategies can begin after accurately profiling technology plan costs. However, even after the plan's costs have been profiled, two additional analyses using the cost analysis framework can be undertaken to examine the sensitivity of costs to variations in (a) any underlying assumptions and (b) the timing of when costs are incurred. The costs of the plan should be examined to see if they can be reduced any further. When the costs of the plan are figured at their irreducible minimum, the second analysis can determine whether the project's costs can be manipulated in ways that reduce the pressure on funding by perhaps shifting costs between different elements of the plan (if the element is easier to fund than another) or from early to late years in the project's life cycle (or vice versa).

In profiling the cost of the plan, several estimates of the expected cost of each item should be made. One of these, of course, will be the current cost of the item. But for most items in the plan there may be certain actions that would lessen these costs. For example, preferential telecommunications tariff rates are becoming more common in states and may be negotiable. Purchasing or leasing the major equipment in the plan might be achieved through consolidated or consortium acquisition arrangements. Some states, for instance, have obtained discounts of 20 percent to 50 percent for hardware and labor costs.21 Technical support for the schools' network may be available from a local university or business, and it may also be possible to piggy-back the schools' Internet connection through that of a university or business. Parent, student, and community volunteers can reduce initial capital outlays by assisting with network installation. Staff training, another large cost, particularly in the more complex plans, can be reduced by conducting sessions during staff's own time, although that practice requires the fullest support of all staff members. The school could also try to obtain donated equipment but should carefully evaluate the equipment’s quality and related maintenance and upgrade costs.

Most of the cost-reduction methods suggested above will require a supporting action plan that must be monitored to ensure that the assumptions on which the plan's final cost estimate is based are realistic and are being met.

In the second type of analysis, the cost analysis framework can be used to evaluate to what extent the demand for funding can be lessened by shifting costs between elements in the plan or from one year to another. For example, project costs can be manipulated to some extent by shifting costs from:

Only when the analyst is satisfied that the plan's costs have been properly profiled and evaluated for additional savings, and that elements in the plan cannot be reconfigured to lessen the demand on funding without affecting its efficacy, can the task of designing the funding strategy begin in earnest.


13 Rothstein. & McKnight (1996). Rothstein and McKnight's overall cost estimates are similar to those produced by McKinsey & Company, but the former's figures tend to be higher for start-up costs and lower for ongoing costs. This seems to be because Rothstein and McKnight allocate more training costs as start-up costs. Rothstein and McKnight's data are used in this section not because their methods are superior to McKinsey's but simply because they are available at the level of detail needed for the type of analysis presented here.]

14 Technology plans can be appraised in other ways, such as the effectiveness of different technologies or the value of investing in school technology versus another type of investment could be evaluated. See, for instance: Council for Educational Development and Research. Plugging-In: Choosing and Using Educational Technology . Oak Brook, Illinois: 1995; and Merrill, D. Evaluation of Educational Technology: What Do We Know and What Can We Know? Washington, DC: The RAND Corporation, May 1995.

15 McKinsey & Company, for example, assumed prices fell by 3 percent per year.

16 U.S. Office of Management and Budget (1992), Guidelines and Discount Rates for Benefit-Cost Analysis of Federal Programs . Washington, DC, Revised Circular A-94, Transmittal Memorandum 64.

17 [ The following discount factors were used to adjust for inflation of 3%:

Yr-1

Yr0

Yrl

Yr2

Yr3

Yr4

Yr5

1.03

1.0

.971

.943

.915

.888

.863


18 McKinsey & Company, for example, assumed prices fell by 3 percent per year.

19 U.S. Office of Management and Budget (1992 ), Guidelines and Discount Rates for Benefit-Cost Analysis of Federal Programs. Washington, DC., Revised Circular A-94, Transmittal Memorandum 64.

20 Viscione J.A. and Roberts, G.S. Contemporary Financial Management, Merrill Publishing Inc. 1987 pp. 96-98.
21 McKinsey & Company (1995).


[Chapter 2 (Part 1 of 2)] Table Of Contents [Chapter 3: (Part 1 of 3)]