The Root of All Money       12

Washington, D.C., is well-stocked with budget wonks who make federal budgeting the stuff of lifelong study. You need not join their ranks in order to enjoy success in the federal information technology market. But an awareness of the seasonal rhythms of the federal fiscal year and the concurrent budgets under consideration within the cycle is vital.

On Any Given First Monday of February

On a typical first Monday of every February, the president delivers to Congress an annual budget request to fund the federal government for the next fiscal year, which starts on October 1. (The budget request is expected to be late when a change of administration occurs.)

Behind the budget request lies at least nine months’ worth of work: federal agencies put together their particular budget proposals; the Office of Management and Budget (OMB) reviewed them and sent them back to agencies, a process called passback; and the request assumed its final shape.

The transfer of responsibility for the budget to Congress is the separation of powers between the branches of government in action. Although Congress legally requires the executive branch—what we’ve been calling in this book “the government”—to prepare an annual budget for Congress to consider, only the legislative branch is constitutionally empowered to pass laws that appropriate public money for federal operations. In short, “the president proposes, Congress disposes.”

Ahead lies a minimum of eight months of congressional maneuvering, as the legislative branch starts the process of passing a budget resolution, arranging hearings, scheduling committee votes (called markups) and floor votes, and finishing reconciliation negotiations between the House and the Senate when they disagree.

At times, rather than bothering with the ordinary course of passing a dozen individual spending bills, Congress puts together an omnibus spending bill, a single bill that makes appropriations for the entire federal government. Sometimes Congress will do so for a handful of agencies at a time, thus making the aggregated bill a “minibus.”

Authorizers and Appropriators

Much of the important action that takes place after the president drops off his budget happens in the House and Senate appropriations committees, which are the committees that control agency spending during the coming fiscal year. They hold the purse strings.

In both chambers, the appropriations committees currently consist of 12 subcommittees, each with jurisdiction over various parts of the government. The charge a subcommittee holds sometimes crosses departmental organizational lines. Defense Department funding, for example, is mostly the jurisdiction of the House and Senate defense appropriations subcommittees. But the DoD construction budget comes from an entirely separate subcommittee, the same one that also appropriates money for the Department of Veterans Affairs.

Subcommittee chairs are referred to as cardinals, and they are indeed among the most powerful members of Congress.

Before appropriators can fund a federal program or function, Congress is supposed to authorize the spending in a legislative process separate from annual budget appropriations. There are two types of authorization laws. There’s enabling legislation, which creates federal agencies and programs or authorizes the drawdown of federal funds without the annual intervention of Congress (i.e., that creates a category of mandatory spending, such as Social Security benefits). And there are authorizing laws that permit appropriations up to a certain level for agencies or programs that the enabling legislation created, assuming they’re not already funded with mandatory dollars. Most federal agencies are entirely funded with such discretionary funding—“discretionary” because without Congress’s passing an appropriations bill, those agencies stop operations.

Authorization laws are typically good for years at a time. The major exception is for defense spending, which Congress subjects to an annual authorization process. (Lending an element of farce to the process is the fact that sometimes the defense authorizing legislation doesn’t become law until after Congress has already approved the annual spending bill.)

Authorization bills assign funding with an air of finality, as if agencies could start spending it tomorrow. In fact, the dollar amounts discussed in nonmandatory spending authorization laws are frameworks, since they merely permit Congress, should it decide to do so, to pass appropriations bills for those programs up to the amount authorized.

Congress also routinely allows authorizations to expire while nonetheless appropriating money for programs anyway. The Congressional Budget Office comes out every January with a report that lists the number of expired authorizing laws and the dollars appropriated to those expired programs. According to the latest report as of this writing, in the last fiscal year, 259 authorization laws had expired for programs for which Congress still appropriated almost $261 billion. Congress can get away with this because authorization laws are not self-enforcing.

All Systems Big and Small

Because the budget process takes so long, even when it’s going according to schedule, government budget planners often directly deal with three fiscal years at any one time. There’s the current fiscal year, which runs from October 1 to September 30. Then there’s the next fiscal year, which is probably under consideration by OMB or Congress, and then there’s the subsequent fiscal year, for which preparations either are about to begin or have already begun. Each year in play contains within it plans and assumptions about future fiscal years.

Let’s pick up the process when agencies start putting together their budget requests, in the spring before the budget is sent to Congress. Requests for major systems typically require individual agency review by an investment review board (IRB). Most program officials for major IT systems must also prepare business cases for OMB review that justify the expense of those systems. The business cases are known as Exhibits 300, or “the 300s.” They’re so named because the language that requires agencies to prepare them is found in Section 300 of OMB Circular A-11. Despite its bland title, this directive is central to the capital planning and investment part of the federal budget preparation process; OMB revises it annually. As we noted in Chapter 1, the intelligence community doesn’t participate in the OMB-led review process; neither do quasi-government organizations such as the U.S. Postal Service, nor the legislative and judicial branches.

There is no hard and fast governmentwide threshold above which a program is considered major. To give some perspective, within the Defense Department, any IT project spending $32 million a year is automatically designated a major automated information system (MAIS) and therefore subject to increased levels of scrutiny. Some agencies will never have a single project that large, but will have investments considered major for them because of the systems’ importance to the mission or function of the agency.¹

See Figure 12-1 for a chart showing how federal spending during fiscal 2011 was divided between major and nonmajor programs, as well as operations and maintenance and development.

As part of the budget preparation process, agencies subject to OMB review also prepare a list of (supposedly) all of the IT efforts they’d like to undertake in the coming fiscal year, a list called the Exhibit 53—so named, yes, after Section 53 of OMB A-11.² At some point after the executive branch hands over the budget request in February, the 53s are posted online.

The Exhibits 53 should be taken with a grain of salt, even setting aside the fact that they’re still subject to final spending amounts approved by Congress. As a list of proposed IT investments, the 53s can be vague. “Multiple projects that serve to enhance the technology foundation, maximize workforce productivity, secure the IT environment, and improve information sharing” is a typical example. And by the end of the fiscal year for which they’re meant to be valid, they’re only somewhat accurate (assuming all the funding came through), at least when it comes to projects that don’t require an Exhibit 300 or that don’t represent static infrastructure spending.

Looking at a neat list of proposed major projects (Exhibits 300) and smaller projects (found in the Exhibits 53, which theoretically list all investments), you might think that at the end of the appropriations process, it’d be easy to find corresponding lists made by Congress with all of the final dollar amounts filled in. Actually, it’s impossible. Although agencies prepare exhibits, they don’t request money that way. Most agencies don’t even have a line item in their budget encompassing the entirety of their IT spend. The IT systems listed in the 53s are mainly subsets of a larger budget line item in which the IT portion is obscure. IT projects typically get funded through money slated for agency programs, or departmental bureaus, or operations and maintenance, or working capital—or some other sort of line item in which the percentage to be spent on IT isn’t specified.³ Really big projects can get their own specific mention in the budget request, but medium- and small-sized projects get wrapped up in larger-purpose line item requests.

The reason OMB created 53s in the first place—and one reason they’re fallible—is that they’re a cross-cut of departmental requests in which IT spending is otherwise embedded and mostly untraceable.

Adding to the confusion is the fact that Congress does not pay much attention to the Exhibits 53. There’s not even consistency in the level of detail at which money is appropriated by one appropriations subcommittee to the next. Some subcommittees give agencies or bureaus a straightforward chunk of money while others make all sorts of stipulations about funding for specific projects.

Big projects, again, tend to receive special attention from Congress, with subcommittees drafting specific language about how much agencies should spend on them or how often agencies should report back to Congress on the projects’ status. Often that language is not in the actual spending bill but in the report that always accompanies each spending bill. Report language technically lacks the force of law and is merely advisory. In real life, it’s optional in the way that parents’ advice about children’s vegetable consumption is optional.

If you look at many spending bills or reports, you will often see that the office of the chief information officer gets assigned a specific funding amount. But even when that’s so, most IT projects aren’t funded through the CIO’s office. Money for the CIO is for the CIO’s function, which in some agencies is just to craft policy. In others, the CIO might be in charge, to a greater or lesser degree, of IT infrastructure for the entire agency and so have funding control over things like the agency network and email system. The more decentralized an agency is, the more likely it is that a CIO’s direct spending power for infrastructure will reach only as far as headquarters’ operations.

The fact that most IT projects are embedded within other line items subject to perhaps vague congressional instruction has a lot to do with how federal agencies change IT priorities. Not that they have total impunity; how much flexibility they have just to take dollars intended for one project and spend them on another still depends a lot on the subcommittee that appropriated the money, the dollar amount involved, and how radical the shuffling is. Agencies might have to make a formal appeal (called a reprogramming request) to the appropriations committee if the deviation from their spending plan is significant. Even if they don’t make a formal request, they might have to notify the committee of their intent to move funds if the amount is large enough to be seen by internal financial controls or exceeds the agency’s comptroller’s authority to reallocate funds without congressional notification.

A considerable amount, but not all, of the money that most agencies spend on IT is one-year money. That is, agencies have until the end of the fiscal year to spend it, and if they don’t, they must send it back. Some is two- or three- or four-year money, but even those dollars have an expiration date. (Congress sometimes appropriates money that agencies can hold onto “until expended,” which agencies appreciate.) Congress takes very seriously attempts to get around funds’ use-by date through accounting gimmickry, so the deadline to spend expiring money by the end of the relevant fiscal year is a real one.

As you may well imagine, program managers are under much pressure to spend to the limits of their authority. Counterproductive and clichéd as it is, programs that fail to spend their full allocated amount will receive less money in the following year, since it’s obvious they don’t need as much. “Use it or lose it,” as program managers say.

Financial officials add to the pressure when they do their best to confiscate unspent money in the latter half of the fiscal year, redirecting it to places where it can be spent within the limits of agency authority. “I always had a financial person who knew that so-and-so has got money left in his postage budget that he can’t spend, and we were ready to pounce on it,” one former senior IT official told us.

Program managers who want to prevent that from happening to them usually have a contingency list of things that would be nice to buy tucked away somewhere in their desk. Such lists are common simply because demand permanently outstrips available funding. It is more likely that a program manager will be able to acquire items from wish lists that contain actionable details, such as well-articulated requirements, a proposed acquisition strategy, and market research, than from lists that lack this information.

Year-end pressure isn’t the only thing goading the government into tactical purchasing.

Program managers and senior agency leadership can change fairly often, particularly in the military, where officers are rotated once every two or three years. The current program manager may have had nothing to do with the planning or the budget of the project he oversees and may be unable to make major changes. But he can push to use, say, operations and maintenance or travel budget money to fund something that shows he was there. This kind of spending also helps explain why so many agencies persist with decades-old technology even while they buy newer stuff—collectively, all these purchases build a fragile ecosystem of hardware and software too expensive to replace wholesale.

There’s an even more off-the-radar type of spending that goes on, too—the type that occurs when agency officials exploit their internal financial control thresholds. Many years ago, a federal agency had a requirement that any information technology expenditure worth more than $50,000 be sent to the chief information officer’s office for review and approval. Along came a clever hardware manufacturer that figured out that by pricing its equipment at $49,800, it could sell boatloads of its stuff to this federal agency—and it did.

Life During a Continuing Resolution

In the past six decades since fiscal 1952, Congress has managed to approve spending bills before the current fiscal year expires exactly three times. More often than not, Congress has failed at some level—subcommittee, committee, the House, the Senate, or entirely as a legislative branch—while attempting to pass spending bills on time. “There’s no such thing as a normal year anymore,” one former government official lamented to us.

Rather than let the federal government grind to a halt for lack of funds, Congress almost always approves a temporary spending bill called a continuing resolution (CR) that allows agencies to continue spending at the rate of last year’s levels until the legislative branch gets its act together.⁴ Often this happens between late December and mid-January. Sometimes Congress will keep the government on a continuing resolution for an entire fiscal year, a state of affairs that federal agencies especially loathe.

Under a CR, new projects aren’t supposed to begin. Agency financial managers also have a tendency to get cranky during partial-year CRs and throttle spending—especially if it becomes apparent that Congress is contemplating a cut.

Congress, although averse to giving up power, is occasionally willing to grant agencies some spending flexibility within the terms of a continuing resolution. But the general condition of life under a CR is one of stasis. And agency managers only get spending authority prorated to the term of the CR, usually on the order of six weeks or so, making it hard to even do capital purchases in support of an ongoing project.

Some agencies take to arguing that a new contract with new vendors for new work is permissible under a CR if the work in question is a continuation of an existing project. How tolerant an agency’s appropriations attorney is of this kind of argument depends greatly on the individual in question and the agency’s relationship to its appropriations subcommittee. Certainly it’s not a line of reasoning to be applied lightly, and it’s definitely not the place of vendors to suggest an agency take it up. The most you can do is hope that senior agency leadership believes an undertaking important enough to engage in a high-stakes discussion on the semantics of “newness.”

Mind you, agencies do continue to buy things and contract for services during a continuing resolution. But the pace of acquisition is slower, and new things that an agency had hoped to do this year might have to wait until next year or could just be dropped from consideration.

Effect of the Budget Cycle on Sales

The particularities of the budget cycle—the timing of the fiscal years, use it or lose it, the likelihood of a continuing resolution for at least the first quarter of the federal fiscal year—obviously have an effect on selling to government.

Get any outstanding deals under negotiation wrapped up before a continuing resolution kicks into force, even if it means sacrificing some margin. A commitment from an agency made during normal times doesn’t necessarily stay valid once a CR comes into effect. One salesman told us a tale of a company holding up a multimillion-dollar software deal with two military offices because one office had exhausted its purchasing budget and was waiting for the next fiscal year to begin—which began under a CR. The company was afraid that by decoupling the sale, it would lose commitment from the office waiting for money. But by waiting in an attempt to bag the Big Kahuna, it endangered the entire sale.

Also, if your company happens to hold a number of annually renewable contracts with a federal agency—whether for a service such as maintenance or an annual software subscription—consider offering the agency an enterprise deal with scheduled renewal in the springtime. You’ll be glad to distance your contracts from the uncertainty of a partial CR, and agencies most likely will, too.

But probably what springs to most people’s minds when thinking about the fiscal year cycle is the procurement ramp-up that runs from July through September, known with some hyperbolical awe as “the fourth-quarter rush.” Agencies, faced with the looming end of the fiscal year on September 30, work feverishly to spend lest they be forced to return funds.

There is money to be made during these months, but the fourth quarter typically is not the bacchanalia about which companies fantasize. Companies that wait to attempt to take advantage of the situation until the fourth quarter actually begins might find that there actually isn’t tons of opportunity, since a goodly percentage of the money spent then will already have been spoken for as agencies maneuvered to ensure that they would not leave unspent dollars on the table. Also, the fourth quarter is a time of high order volume, a phenomenon that decreases the average value of acquisitions.

Typically, a majority of fourth-quarter spending is done through standing contract vehicles rather than through full and open competition simply because of the lead time the latter kind of procurement necessitates. Agencies also rediscover small businesses during the latter half of the fiscal year because competition standards for socioeconomically favored firms are less stringent.

If you want to be a meaningful part of an agency’s fourth-quarter rush, the time to go to program managers is when they’re looking at their spending rates and reexamining their current plans. Agencies start looking for low burn rates among program offices starting in the second half of the fiscal year, so the time to discuss these matters is before then, starting around January. That’s not to say that you shouldn’t be meeting with feds to discuss their needs during the rest of the year—you should, and should be doing so constantly.

The bona fide needs rule

Congress, as we’ve seen, is vigilant about prohibiting agencies from spending money after it has expired. That also means that agencies must have a legitimate need for the products or services they’re buying within the fiscal year of the funds they’re spending. Were agencies to spend current fiscal year dollars on future needs, that would be the same as spending money after its expiration date, says Congress.

This is known as the bona fide needs rule. Agencies must have a sincere (bona fide) reason for spending current fiscal year money. The sincerity (the bona fideness) is measured by an agency’s current, not future, needs.

For federal buyers, satisfying the bona fide requirement toward the end of a fiscal year can sometimes be tricky, especially if the goods they order can’t be delivered until after the fiscal year has turned. Such orders are permissible but can draw the suspicion of inspectors general and other watchdogs.

One way civil servants get year-end purchases through the bona fide gate is to characterize them as resupply or part of a larger plan. What constitutes a “resupply” purchase may be somewhat ambiguous, but it’s nonetheless a concept that generally permits agencies to buy new things on the grounds that old things have been consumed or used up during the current fiscal year, creating a current need for replacements (although the value of the replacements must not exceed the value of the current need).

Bona fide is something that federal officials, more than private-sector sellers, need worry about. However, if you’re trying to make a significant sale toward the end of the fiscal year, it’s a good idea to keep the issue at the forefront of your buyers’ minds. The danger is that the issue of bona fide need could be overlooked, and it would be a shame for the sale to fall through at the 11th hour on those grounds.

Severable vs. nonseverable services

How is it, then, that services contracts are able to cross fiscal years without running into funding problems? The explanation can depend on whether the service is severable or nonseverable.

A severable service is a task that realizes the value of the contract incrementally, maybe on a daily basis. Help desk support is one example. The value of the service doesn’t depend on a single outcome, but is realized each time a help ticket is resolved.

Nonseverable services depend on a single, discrete outcome—for instance, a research project. There’s no incremental value to a day’s worth of experiments, only in a fully concluded study. Agencies are allowed to fully fund nonseverable services contracts with current fiscal year appropriations, even when their outcome will occur in future fiscal years. After all, the need is current—it’s just the outcome that’s deferred. Training is classified as a nonseverable service; in fact, training classes need not even begin in the fiscal year during which a training contract is let, so long as they begin soon thereafter. Annual subscriptions also are nonseverable.

All agencies except NASA are also allowed to fund severable services contracts with current fiscal year funds for a period of up to 12 months, even if that 12-month period crosses the federal fiscal year.⁵ Before 1995, the law stated that all agencies had to pay for severable services with appropriations for the fiscal year in which the services were actually performed. As services contracting really took off, this proved to be a bigger headache than it was worth, and Congress relented.

Final Note

Federal fiscal years and the budgets that underpin them are artificial creations, unlike, say, the seasons of an actual year. But we hope we’ve demonstrated that the timing of the federal fiscal year can be no less important to your business than the arrival of summer is to a beachside ice cream shop. Don’t expect boom times during a continuing resolution or to start selling a major new product in September. Do learn to await the public kickoff of the federal budget in February with anticipation, since that’s when you can get a look at what the federal government considers to be its priorities. Likewise, make sure to catch up with the final numbers once Congress approves them (even if doing so deepens your level of cynicism about our democracy). But mainly, make sure you coordinate your activities with the season of the fiscal year.

ENDNOTES

1. DoD systems that will cost $126 million for all program acquisition costs or $378 million for total life cycle costs, including operation and maintenance, also count as MAIS; all dollar figures are in fiscal year 2000 constant dollars (10 USC 144a).

2. In recent years the requirement has shifted to section 51.18 of A-11, but the name goes on.

3. Operations and maintenance money, despite sounding like a budget line item that does nothing other than maintain a steady state, in fact is the source of many, albeit typically smaller, new IT projects.

4. Almost always. In 1995 and 1996, President Bill Clinton vetoed two continuing resolutions rather than approve substantial cuts and legislative changes the Republican-led Congress had placed in the temporary spending bills, which traditionally are not places to make new policy. As a result, the federal government actually did run out of money twice in the resulting showdown.

5. FAR 37.106(b).