The public sector makes huge investments in computers and telecommunications -- roughly $70 billion per year in direct costs, and an estimated six times that amount in indirect costs -- yet we continue to see major problems in how government handles technology. Why?
WHAT'S THE PROBLEM?
We clearly see that IT has become an agent of change, constantly confronting us with significant choices regarding what we want to do and become. Unfortunately, we're not handling these choices as well as we could. More than half of all reengineering projects have reportedly failed, even in the private sector. In the public sector, technology-related tales of woe are even more endemic. The traditional budget process -- with its emphasis on year-at-a-time, bottom-up, incremental modifications -- tends to overlook innovative IT projects that cross program and budget year boundaries. What can we do to help government do a better job of finding and funding good IT?
To answer this question, we first looked at the views of those at the frontlines of government IT endeavors. Through surveys, we asked them what they are doing, what value they see being created (or not being created), and how hard it is for them to implement their initiatives. Collectively, they considered the highest-value initiatives to be infrastructure and cross-boundary integration, and the easiest to be educating practitioners and using capital budgeting and other multi-year vehicles to fund the creation of multi-year information assets (see sidebar). Perhaps most interesting, they identified two high-value options -- performance contracting and R&D/learning investments -- that are also relatively easy to implement but not yet given high priority.
These are potentially powerful ideas, but what can you do to act on them?
KEY ACTORS HAVEN'T CONSIDERED IT "THEIR" PROBLEM
At some point, after you recognize a problem and identify possible solutions, you need to generate collective action -- a concerted effort toward change. Establishing a task force or team to reform particular government processes can help an agency remove systemic and procedural barriers to better IT funding. The three most important processes to target are the budget process, the IT strategic planning and oversight process, and the IT project implementation process. Improvements in any of the three will help, but improvements to all three will be even better. Budget directors, chief information officers (CIOs) and department heads all have important roles to play.
FOR THE BUDGET DIRECTORS ...
Changing the budget process is often the most powerful way to change how government handles something -- IT issues as well as others. The "team" you will need for budget reform will typically include the budget director, the elected or appointed official the budget director reports to (mayor, department head, governor) and -- at least to some extent -- the budget overseers, especially the legislative overseers who receive, modify and eventually approve the budget.
If the budget director agrees with you, you can usually move forward without much fuss. And, of course, if you are the budget director, the budget reform agenda must be your own. You should ensure that budget preparation gives good guidance and attention to IT-related issues including:
* Infrastructure and standardization;
* "Cross-boundary" data sharing and service integration;
* R&D and learning investments; and
* Funding via nontraditional, nontax levy appropriations (capital funds, revolving funds, user charges, and/or performance contracts).
Budget reform will require you to educate and train every participant in the new concepts and procedures to be used. The focus should not be on technology per se, but on the business results that technology can create. You should be sure that budget proposals requiring IT-related organizational change are made and reviewed primarily by line managers, not technology staff; this will usually require you to prepare the line managers for new IT leadership responsibilities. You should see that well-established, centrally-provided IT services are funded via charge-back fees rather than direct allocations, as this gives power and responsibility to departmental line managers, especially when departments have the option of taking their business to outside providers. On the other hand, for experimental, new and not-well-established services, you should implement revolving funds or other overhead accounts that don't place too much of the funding burden on early adopters.
For these and other reforms, you may want to organize your efforts on a basis broader than IT issues alone. Much of what you will need can be accomplished via reforms focused on themes, such as performance measurement, work process improvement, total quality management, outsourcing, the introduction of strategic planning and program analysis into the budget process, and so forth. For any of these efforts to be successful, expect to assign a team or task force of your best staff, supported by consultants and other outsiders, for three to six months or more before the first season in which you prepare your budget under the new procedures. If you have not already done so, you will probably want to develop an analytic staff unit to explore a range of annually-defined issues on a deeper basis than traditionally can be provided by your departmentally-oriented budget staff.
Results of the Practitioner Survey
Mean Scores for 12 Initiatives to Improve IT Funding in Government
Item Social Value
Better Participation initiatives:
Line leadership 3.63
Oversight involvement 3.39
Better Portfolio initiatives:
R&D and learning investments 3.99
Infrastructure investments 4.36
Cross-boundary investments 4.18
Non-Tax Funding initiatives:
Revolving funds 3.68
Capital funds 3.47
User charges 3.43
Performance contracts 4.13
1 = low, 10 = high
FOR THE CIOS ...
If you are the CIO, the IT planning and oversight agenda will largely be your own. To begin, you may want to start with the informal planning process -- the many conversations, often one-on-one, that help define issues and establish a pattern of networking to support the work of the group. If you were chosen as CIO largely because of your interpersonal and networking skills, use them. On the other hand, if you were chosen largely for your technical or other skills, you will need to develop your communications and leadership abilities to the greatest degree possible (although, in general, this will be more difficult than developing your understanding of the technical concepts required to be a good CIO).
In most jurisdictions beyond a certain size -- say 500 or more employees -- important elements of the planning process are likely to depend on written analyses and plans. In these settings, you will want to institute an IT planning process to address issues that include:
* Standards for the networks, data and skills to be shared as infrastructure throughout the enterprise;
* Organizational responsibilities for IT, including reporting relationships and control over activities, such as applications development, systems operation, maintenance and training;
* Design and deployment of electronically-delivered services, especially opportunities for service integration and self-service; and
* Coordination with the private sector on electronic commerce and attracting good jobs.
The planning process should be overseen by line managers, senior civil servants, and elected leaders and their appointees, not just by technology managers. While IT planning should be distinct from budget preparation in most settings, it must maintain budgetary legitimacy at least to the extent that proposals rejected in the IT strategic plan do not receive funding in the regular budget.
To integrate IT capabilities and priorities appropriately into the organization's culture, you need ongoing oversight that, like the strategic planning process, is controlled by IT-knowledgeable general managers rather than by technologists. In most settings, you will need a steering committee primarily of government insiders to meet about every three months to keep IT activities on track. In addition, you will want an advisory board of outsiders to keep you connected to the vendor community, academia, other levels of government and so forth.
FOR THE DEPARTMENT HEADS ...
While most IT initiatives start with the budget process and the IT planning process, they ultimately succeed or fail as implementation projects. And for IT implementation projects -- the trench warfare of IT progress -- success depends more on department heads and line managers than on budget directors or CIOs.
If you're a department head or line manager whose unit depends on IT -- whose doesn't? -- the short list of essentials for IT implementation includes:
* Get line management (and probably yourself) deeply involved at the project level; many projects will need a general manager rather than a technology manager as the authority in charge of daily project activities.
* Make a business case for funding based on business outputs (not just technology outputs), with justification that includes benchmarking against outside organizations (where possible) and an open assessment of the confusion and conflict that must be resolved (always).
* Seek nontraditional funding -- capital funding, revolving funds, user charges, performance contracts -- if needed and justifiable.
* Provide solid education for all project participants, including overseers.
Real progress in helping government improve how it finds and funds IT will require engaging overseers, the press and the public in what has been referred to as "the conversation" of the political process. We need to support government in a disciplined way as it tries to understand and keep up with new information technologies. We don't want to spend blindly, be soft or cave in to narrow interests -- of vendors or others. But we do need to find better ways to authorize government to experiment, to change and to "fail forward." To do this, we will need to pay more attention -- both as citizens and as leaders -- to holding government accountable for an energetic learning portfolio. This will present a significant challenge, of course, given the ambiguities and conflicts involved with learning and change. But we have begun over the past decade to improve how we learn about and respond to IT issues and opportunities, and we must continue.
In a world characterized by much better access to information and increasing interdependency, how governments handle information technology will be important, even critical. We know from our past failures that future progress will be fitful and risky. But we also know it will be essential to push ahead and to do so with as much thoughtfulness and discipline as possible.
Finding and funding good IT in government has become a necessity, not a luxury -- so now is the time to act. The Strategic Computing Program at Harvard has long been interested in the ways that agencies face the challenges presented by the IT budgeting problem. As you begin crafting your own strategy, keep us posted on your progress.
Jerry Mechling is director of Strategic Computing and Telecommunications in the Public Sector at Harvard University's John F. Kennedy School of Government. Victoria Sweeney is a research affiliate with the same program. *
This is the final in a series of four articles drawn from the forthcoming Government Technology publication, "Finding and Funding IT Initiatives in the Public Sector: A Guidebook." The book is designed to pull together the information and advice that a task force charged with solving IT funding problems needs to organize its work.
April Table of Contents