Launch Vouchers Offer New Space Research Opportunities

On April 28, 1989, with just 31 seconds left to go, the nation's first major planetary space mission in ten years was put on hold. An electrical problem in the space shuttle's main fuel pump had delayed yet again the Magellan mission to Venus. While another delay in a mission that was already six years late was disappointing—particularly to some of the country's most renowned planetary scientists, who were among the tens of thousands gathered for the lift off—even more disturbing was the narrow window of opportunity for launching Magellan. Venus and the earth would be properly aligned for only 29 days; otherwise, the $300 million Magellan project would have to go back into storage until at least January 1991.

Such are the onerous demands of space science and a temperamental shuttle program. These demands had led to a backlog of unlaunched missions even before the three-year delay in space transportation following the accident of the shuttle Challenger in January 1986, and the accident served to make the problem worse. Now, in what one expert calls an ongoing game of musical space shuttles, the scramble to launch Magellan (which finally lifted off, four days late) has left in limbo the takeoff of another long-overdue mission, the Hubble Space Telescope. In turn the retrieval of a carrier called the Long-Duration Exposure Facility (LDEF) has also been delayed. That facility, containing automated experiments in materials science, has been in space since 1984. If not retrieved soon, LDEF is in danger of falling back to earth.

These large science missions, together with thirty others, are not the only players in the musical shuttle game. Also scrambling for launch are over 100 smaller space science research payloads. All told, the delays for large and small payloads currently extend to six years or more.

These delays bring huge costs. Storing payloads and maintaining their flight readiness can total up to $100 million a year for each large payload and $5 million or so for smaller ones. Indirect costs are more difficult to measure but potentially are as important. They are likely to include attrition among space scientists in universities and industry, as well as declining enrollments of new students in space science. To be sure, the field of space science requires patience—even on-time missions can take eight years or more before a planetary destination is reached—but an additional four years or more of delay in getting off the ground is bound to discourage space research.

Vouchers to the rescue?

How to increase opportunities to get to space was addressed in national space policy announced in February 1988. The policy called for consideration of a radically new approach to space access, "space transportation vouchers." As envisioned by the policy, vouchers would permit research missions currently queued up for launch on the space shuttle to purchase alternative, commercial U.S. launch services. These services are represented by a variety of conventional unmanned rockets that, in years past, have routinely launched communications satellites and other payloads for private industry and government.

Although the policy does not detail how vouchers might work, they presumably would operate much like vouchers that have been used in U.S. federal housing programs. Certificates issued and financially backed by government would be given to researchers for redemption on any mode of space transportation—the shuttle as well as unmanned launchers. Researchers in private industry, government, and universities could be eligible, and voucher-supported research topics could run the gamut from materials, life, and earth sciences to engineering research and plasma physics. Vouchers thus support the tradition that space research should be publicly funded at least in part, but bring the possible advantages of a market like mechanism to the process of allocating these federal research funds.

The voucher proposal contrasts markedly with the present system. Payloads now fly in an order determined by centralized administrative decisions. A shuttle flight is first scheduled around a government-sponsored large payload (like Magellan) that takes up much of the shuttle's carrying capacity (about 48,000 pounds), and smaller payloads (ranging from 60 to 10,000 pounds) are then added on a first come, first served space-available basis. Examples of smaller payloads range from an experiment to evaluate the interaction of oxygen with various materials, to so-called Get-Away-Special canisters ("GAS cans," ranging from 60 to 200 pounds; see figure 1) and "Hitchhikers" (750 to 1,200 pound canisters), which might contain anything from an ant farm to sensors for studying ultraviolet emission. These self-contained payloads are predominantly designed by researchers in industry, although some (such as the ant farm) are projects sponsored by industry and designed by post-secondary and even elementary and secondary school students.

Smaller payloads such as these, then, are scheduled jointly with large payloads. Shuttle delays thus ripple throughout the space science community. And further delays seem inevitable given constantly moving shuttle schedules and cutbacks in future shuttle flight rates.

Vouchers might alleviate these problems for several reasons. First, by allowing researchers to make use of conventional rockets rather than requiring them to use the shuttle, vouchers could increase the supply of space transportation. This flexibility would relieve schedule pressure on the shuttle—whose managers could then focus on launching payloads that require human interaction—and reduce the backlog of missions sooner than exclusive reliance on the shuttle might permit. Second, vouchers might allow the realization of the benefits from space research earlier than is likely to be possible by way of the shuttle. In turn, by demonstrating demand for space access, vouchers might stimulate the supply and diversity of commercial launchers, and, in the future, of payload return vehicles.

Other potential benefits of vouchers include two by-products of expanding near-term flight opportunities for space research. One would be a gain in experience to better inform the highly contentious debate over funding U.S. space science in general and the proposed space station in particular.

An additional by-product might be greater participation in space activity by industry and non-NASA government agencies. By offering more certain and timely access to space, vouchers might alleviate the financial and technical uncertainty that presently inhibits funding of space science by industry—investors generally saying "no buy if it won't fly soon." In addition to increasing industry participation, voucher-facilitated access to space might spur space research by others in government. A 1988 National Research Council report, "Industrial Applications of the Microgravity Environment," notes that government offices possibly interested in space science research include the National Oceanic and Atmospheric Administration, the National Institute of Standards and Technology, the Department of Energy, and the National Institutes of Health.

Possible cost savings

Even if these by-products did not materialize, a good case can be made for vouchers on the basis of their cost savings in comparison with the current system. Consider first the cost of vouchers. A full-scale voucher program that issues vouchers to all payloads requesting flight dates between now and 1995 could cost about $4 billion. This estimate would be significantly less if some payloads could be modified so as not to require return to earth. By comparison, space transportation services provided solely by the shuttle are estimated to cost on the order of $3 to $6 billion. It should be emphasized that these estimates measure the cost of the shuttle program at real resource costs. The reported cost of a shuttle flight typically includes only the cost of fuel and other expendable items. Yet depreciation of fixed facilities (such as launch pads and reusable orbiters) and other capital costs, in addition to the cost of expendable items, constitute a truer measure of resource costs. These costs must be taken into account in order to see the full social cost of the shuttle.

Thus there is a good chance that vouchers could save up to $2 billion if the costs of the shuttle program are at the high end of this estimate. Even if shuttle costs are on the order of $3 billion, however, vouchers could bring cost savings if some payloads do not need to be returned to earth. Moreover, when delay costs are taken into account, a one-year delay just in the larger planetary and astronomy missions could readily justify any cost difference between the voucher program and the cost of shuttle launches. The indirect costs of delay, such as declining interest in space research careers, would further support the cost-advantages of vouchers.

Promising institutional change

Aside from potential cost savings, vouchers would also bring a crucial change in approach to managing two areas: space transportation and space research. The present administrative approach sharply separates these activities; they are managed and budgeted in different NASA offices. Such a division of responsibilities leads to a host of problems: NASA task forces note that these include inefficient use of resources, wasteful competition for resources, and ambiguous and conflicting goals. For example, consider the transportation-related questions a space research experimenter faces in designing a payload: Should the payload be automated, or make use of human interaction? What should be the on-orbit duration of the experiment? Should the payload be returned to earth or should data be collected by computer and relayed home electronically? These transportation concerns represent expensive engineering tradeoffs leading ultimately to choice between use of the shuttle and conventional rockets. Such decisions are now made without full information about the relative cost of these tradeoffs.

By allowing researchers a choice between transportation modes, vouchers could force a closer coupling of the budgetary and cost impacts of payloads and space transportation. Just how this coupling would take place would depend on the design of the voucher.

Possible disadvantages

Vouchers, however, may not be without their own pitfalls. For instance, how quickly can payloads now configured for shuttle launch be refitted for alternative vehicles? Is there a sufficient supply of alternative vehicles to meet voucher demand? Would a one-time voucher program, as envisioned by the national space policy, be adequate to spur demand and supply? Are the high reentry forces for return vehicles other than the shuttle an insurmountable technical difficulty?

With these problems in mind, there are at least three alternative voucher programs worth considering (summarized in table 1). Under one program, vouchers would be issued for a standard face value that was less than projected total transportation costs. The difference would be made up by copayments from payload sponsors. Copayments would ensure against incentives for researchers to overstate transportation and payload return requirements if there were no penalty for so doing (the penalty is analogous to copayment in medical insurance). Copayments could, for instance, be required for the return vehicle, thus forcing payload owners to better assess experiment design alternatives (such as automating their payload).

A second design alternative would be "cashable" transportation vouchers. Cashable vouchers would be valued at the estimated cost of unmanned transportation, but would include a provision under which recipients could keep the difference between estimated and actual costs if the latter were lower, provided that difference is allocated to space research. The advantages of cashable vouchers are that researchers would be encouraged to search for low-cost transportation, and that the burden on government to guess transportation costs precisely would be reduced. In addition, if the transportation cost savings were divided between the researcher and the transportation supplier, suppliers as well as researchers would have incentives to lower costs.

A third alternative would be to issue vouchers for an entire space research project rather than for its transportation component only. Such space research vouchers could be funded from space research budgets augmented to include transportation. They would provide space scientists with the greatest degree of choice in all aspects of the research effort: in searching for low-cost transportation, in designing payloads with transportation requirements and costs in mind, and in allocating the budget among transportation, payload design, ancillary ground-based lab facilities, and even professional staffing.

Of these alternatives, research vouchers would best reduce the sharp discontinuity now existing between research and transportation. Research vouchers would also align space research (and the management of the project budget) more closely with the process of research in other, non-space fields, where grants are awarded to a project as a whole. Thus space science might be better able to compete for talent with other research fields, and maybe the long line of space access would shorten.

Any of these design alternatives should allocate the burden of space transportation and research risk between the government (as the founder of space transportation) and the commercial supplier of the transportation. The economics of risk-sharing offers some guidance. It suggests that the burden should rest with the supplier, given the greater information the supplier has about actual costs and risk and the difficulty of monitoring them.

As a step toward designing a voucher system, a pilot program could be undertaken to test and evaluate different designs on a small scale. By analogy with the evolution of housing vouchers, Congress could direct NASA to establish an experimental space transportation allowance program. It could specify voucher payment formulas, target dates for completion, and conditions for eligibility. The program could be undertaken at existing Centers for the Commercial Development of Space—collaborative university and industry programs set up by NASA to promote space science research.

Like housing vouchers, space vouchers would require a multiyear budgetary commitment. The government might also need to finance part of the costs of any major investment in unmanned launch facilities or return vehicles necessary to accommodate space science demand. Based on the estimated costs of vouchers, however, even this investment would be likely to provide space transportation at lower cost than is presently incurred by the shuttle program. Thus all signs suggest "go" in launching a pilot voucher program.