IT's Capital Fault Lines

By Marcia Gulesian

(Back to article)

Capital budgeting is one of the most important tasks faced by CIOs, financial managers, and their staffs. Determining the correct level of IT capital spending requires insights that only a CIO can provide together with other insights that only a financial officer can provide.

First, a firm’s capital budgeting decisions define its strategic direction, because moves into new products, services, or markets must be preceded by capital expenditures.

Second, the results of capital budgeting decisions continue for many years, reducing flexibility.

And, third, poor capital budgeting can have serious financial consequences. If the firm invests too much, it will incur unnecessarily high depreciation and other expenses.

On the other hand, if it does not invest enough, its equipment and/or software may not be sufficiently modern to enable it to produce competitively. Also, if it has inadequate capacity, it may lose market share to rival firms, and regaining lost customers requires heavy selling expenses, price reductions, or product improvements, all of which are costly.

A recent news headline serves to illustrate the inevitable downstream cost to one company for not having invested enough capital in its servers and network in a timely fashion: "Penny Wise, Pound Foolish".

This April, on the morning following the deadline for filing U.S. federal income tax returns, the Internal Revenue Service said it would not penalize TurboTax and ProSeries users whose 11th-hour electronic returns were delayed by Intuit's over-loaded servers.

A record number of returns from both individual taxpayers and accountants started causing delays in customers receiving online confirmation that their tax returns were submitted successfully.

As the midnight filing deadline approached, the problem got worse. During times of peak demand, Intuit was processing 50-to-60 returns per second.

Fortunately for the affected taxpayers, the IRS announced that filers who encountered delays with Intuit's servers will not be penalized if their returns rolled in late. The reprieve also extends to people who tried to file in the hours before the deadline but could not.

Had Intuit invested more capital to upgrade it servers and network, it would not now be facing, post priori, operating expenses needed to restore customer confidence in its services.

Capital vis-à-vis Operational Budgets

Capital spending on, for example, new hardware or application server software, and operating spending on, say, salaries and other public relations expenses, are different and yet, as suggested in the previous sentence, related.

Often the time period for spending and generally the means of financing (for example, borrowing for capital spending vs. current revenues for operating spending) are different.

Though they are different types of budgets, the capital and operating budgets are interconnected in many ways, foremost by annual service of debt or other obligations that finance the capital plan.

Annual debt service is usually included in and paid from the operating budget. In addition, many capital assets, such as computer hardware or software, require annual operating expenses for operation and maintenance, among other items. In some cases, capital spending in a given year may be financed by using current revenues (pay-as-you-go financed).

Carefully developed capital projects can also save operating expenses, by creating more efficient ways to provide services, or by introducing energy-saving measures, for example. The impact of capital expenditures in total on the operating budget should be carefully considered before capital budgets are made.

In advance of committing to a given level of capital investment, the debt service needed to support borrowing (or the amount of cash earmarked for dividends) needs to be calculated as is the need for new revenues.

Thus, the capital budget is constrained by revenues available to pay for it. New or available revenues to pay debt service define debt affordability, which drives the size of the new financing program.

The concept of debt affordability, in particular, takes into account the long-term effects of debt repayment on operating budgets. And, debt service obligations, without new funding, will compete with the undertaking of new projects.

A debt affordability policy is needed to maintain a firm’s good credit rating and, thereby hold down its cost of capital (discussed below).

Project Classifications

Analyzing capital expenditure proposals is not a costless operation. For certain types of projects, a relatively detailed analysis may be warranted. For others, simpler procedures should be used.

Accordingly, firms generally categorize projects and then analyze those in each category somewhat differently:

1. Replacement: maintenance of business. Replacement of worn-out or damaged equipment is necessary if the firm is to continue in business.

The only issues here are (a) should this operation be continued and (b) should we continue to use the same production processes? If the answers are yes, maintenance decisions are normally made without an elaborate decision process.

2. Replacement: cost reduction. These projects lower the costs of labor, materials, and other inputs such as electricity by replacing serviceable but less efficient equipment. These decisions are discretionary, and require a detailed analysis.

3. Expansion of existing products or markets. Expenditures to increase capacity or output of existing systems in markets now being served are included here.

These decisions are more complex because they require an explicit forecast of growth in demand, so a more detailed analysis is required. (The Intuit story presented above falls into this category.) Also, the final decision is generally made at a higher level within the firm.

4. Expansion into new products or markets. These projects involve strategic decisions that could change the fundamental nature of the business, and they normally require the expenditure of large sums with delayed paybacks.

Invariably, a detailed analysis is required, and the final decision is generally made at the very top—by the board of directors as a part of the firm’s strategic plan.

5. Safety and/or environmental projects. Expenditures necessary to comply with government orders, labor agreements, or insurance policy terms are called mandatory investments, and they often involve non-revenue producing projects. How they are handled depends on their size, with small ones being treated much like the category 1 projects described above.

6. Research and development. The expected cash flows from R&D are often too uncertain to warrant a standard discounted cash flow (DCF) analysis. (Please see my article on Developer.com for a detailed discussion of DCF).

Instead, decision tree analysis and the real options approach are often used. (Please see my article on Developer.com for a detailed discussion of decision tree analysis and the real options.)

7. Long-term contracts. Companies often make long-term contractual arrangements to provide products or services to specific customers. For example, IBM has signed agreements to handle computer services for other companies for periods of five-to-10 years.

There may or may not be much up-front investment, but costs and revenues will accrue over multiple years, and a DCF analysis should be performed before the contract is signed.

The Cost of Capital

The cost of capital is affected by a number of factors. Some are beyond the firm’s control, but others are influenced by its financing and investment policies.

The three most important factors that are beyond a firm’s direct control are: (1) the level of interest rates, (2) the market risk premium, and (3) tax rates.

A firm can control its cost of capital through: (1) its capital structure policy, (2) its dividend policy, and (3) its investment (capital budgeting) policy.

The cost of capital is beyond the scope of this article and beyond the responsibility of the CIO. At the same time, understanding the operational consequences of under-investing or unnecessarily investing in the firm’s IT infrastructure can only be estimated by the CIO and his/her staff.

Thus, the level of IT capital investment needs to be based on the judgments of both the CIO and his/her finance counterpart.

Marcia Gulesian has served as software developer, project manager, CTO, and CIO over an eighteen-year career. She is author of well more than 100 feature articles on IT, its economics, and its management, many of which appear on CIO Update.