written by Strategic Institute Staff
“The more things change, the more they stay the same.” Early on, OT authority and the first agreements were misunderstood; subject to poor reporting that resulted in false assumptions which were then accepted as accurate. While senior leaders gradually came to see the advantages of OTs, many mid-level bureaucrats ignored them or resisted implementation. In short, the early environment regarding OTs is not much different today.
The First OTA
The original OT authority (10 U.S.C. 2371) was vested solely in the Defense Advanced Research Projects Agency (DARPA). For the first thirty years of its existence, DARPA did not have an internal contracting office, doing all its work through external contracting agents. ARPA Orders were sent to various (mainly DOD) contracting offices for execution, which was in accord with DARPA’s desire to retain a lean, low overhead, and flat organizational structure. External agents were selected because they had an interest in DARPA developed technology and could be a conduit for transition of a successful project into operational use. By the late 1980’s, DARPA had acquired a contracting office and a general counsel. However, it continued to execute most of its projects through external agents.
DARPA officially entered into its first Other Transactions Agreement in the spring of 1990 with Gazelle Microcircuits (Gazelle), a venture capital supported company. Gazelle was tasked to develop high-speed gallium arsenide (GaAs) data communication components, and was awarded a $4 million, 12-month OT agreement. DARPA’s involvement provided Gazelle with “development, design, production engineering, and working capital to develop and bring to market high-speed data communication GaAs components, electronic modules or subsystems, and application development tools.” In return, DARPA retained “access to research and development results; certain rights in data patents; and in the case of technology developments that resulted in commercially marketable products, a fair return on it’s investment and discounts for Government purchases of such products.” This was the first time that a “dual-use” agreement with ROI had ever been forged between the government and a commercial company.
Some backstory: DARPA had previously played a MAJOR role in the establishment of a digital GaAs manufacturing capability because the technology offered defense systems high-speed operation with low power consumption. Due (in-part) to its early efforts, U.S. companies were manufacturing the world’s most complex GaAs integrated circuits with the highest yields. DARPA’s goal with the new GaAs program was to “implement this technology and exploit the U.S. manufacturing capability for critical defense needs.”
It is important to paint a picture of the temperature of American politics at the time. This was the early post-cold war era. America’s biggest competitors were Japan and Europe; military tensions between countries were starting to ease, however economic competitiveness, especially in key high-tech areas, was now the important factor in the “global balance of power.” The Bush (HW) Administration strongly preferred the defense industry to stick with weapons, and not invest in high-technology. An internal war over how America would move forward was quietly raging behind the scenes.
With the OT agreement in place, two press statements were released in early April. From the Pentagon: “DARPA Signs Agreement for Gallium Arsenide Work;” from the Washington Post (WP): “Pentagon Agency Invests $4 Million in Chip Firm.” This seemingly insignificant WP headline pushed the Gazelle agreement to the forefront of American politics. The headline received all the attention, while the actual content and facts were disregarded. The agreement was misconstrued to imply that DARPA had invested capital in Gazelle and was now operating as an equity partner of the small business, which was simply NOT TRUE!
So, what was the problem? There was no real problem. The only problem existed in the narrative that was portrayed within the media. The term Investment was a stumbling block for accurate reporting at the time. The idea that DARPA had actively invested in and encouraged developing advanced commercial technologies, like sophisticated semiconductors, was highly disliked by some high Administration officials, who strongly believed that “technologies were better developed with funds from business and industry.” The real sticking point was that the deal allowed Gazelle to deflect investment from a Japanese company. The Pentagon (who preferred to buy technology from the Japanese after development) could not understand why the government would invest, when a private company was willing to put the money in.
At the time, it was a popular idea on Capitol Hill (much to the dismay of the then current administration) that the government should help support and invest in certain key industries, like computer chips and advance materials, because of its impact on the U.S. industrial strength. Where had the government done this before? In the prior generation, they built an entire national transportation system (roadways, bridges, etc.) for railways and automobiles. It’s ironic that the administration was so reluctant to invest in the “information highway.”
The Good News
Despite unhappiness with adverse publicity, senior DOD leadership soon discovered that the agreement had been reviewed and approved in advance by the Department’s senior acquisition officials. When Deputy Secretary of Defense, Donald Atwood, testified before the Senate Armed Services Committee, he endorsed the agreement saying it would remain in effect: “I think the authority is appropriate…the particular project they undertook…is something we ought to foster…I have no problem with that at all.” The agreement was successfully completed, and Gazelle microchips made their way into a wide variety of DOD development and production systems, along with commercial products. DARPA received returns from the sale of commercial products which were credited to the special account authorized by the statute.
Before the dust settled, DARPA was directed to have future agreements reviewed by the director of defense procurement. This nonvalue-added process, which discouraged and delayed execution of agreements, lasted for two years. At the end of 1992, the role of the director of defense procurement in DARPA’s OT process was terminated and DARPA was free to make innovative use of its OT authority. One of the first projects undertaken was a cost-shared consortium agreement, whose payoff is still being seen today in the ceramic matrix composites that contribute to the high thrust to weight ratio of current gas turbine engines. The ceramic fiber consortium, in support of the Integrated High-Performance Turbine Engine Technology (IHPTET) program (Figure 1), was also instrumental in garnering Congressional support for expanding the use of consortia to engage in precompetitive technology developments and other ways of partnering between government and industry and among industrial companies.
Figure 1: Funding agreement for Integrated High-Performance Turbine Engine Technology (IHPTET)
Interestingly, the idea for this arrangement did not arise exclusively at DARPA, but with the Air Force (Wright Materials Lab) at a time when the AF did not have OT authority. One AF idea that was not carried forward in this arrangement was for the jet engine companies to form a non-profit entity to receive government funds (because the AF contemplated awarding a grant). When given the option, the companies all preferred not to create a non-profit entity. Funding came from the sources indicated. The private companies did not all contribute the same amounts and their management vote was weighted according to their contribution. The non-profit integrating sub-contractor was strictly a fee for service administrator and did not extract a percentage burden of the pass through of funds (more funds to research rather than overhead).
One key aspect of this was that the “usual suspects” that did research in high temperature ceramic fibers, were required to partner with a materials manufacturer and create a product that could actually be produced and go into a jet engine application. While decisions on what research to fund were made openly by the company’s management committee, the actual use any company might make of the results remain proprietary.
This is another key point. In OTs, an IP regime can be created that is partially open and partially proprietary. For example, a distributed medical information system in which the backbone, system reference architecture, as well as all interface specifications were developed openly and jointly, but each consortium member also contributed a proprietary/black box node to the system. In future generations, those black boxes could be seamlessly replaced by new or even competing versions. The IHPTET performers, who were chosen via a competition conducted by Integrating Sub Contractor (but with decisions by the companies), did not necessarily retain any IPR in their work. The basic program that this consortium supported, Integrated High Performance Turbine Engine Technology, did not reach all its goals, but ceramic matrix composites are a key component of current military jet engines.
With the backdrop of the earlier SEMATECH cost-shared semiconductor manufacturing technology program and examples such as the IHPTET joint funding agreement, DARPA demonstrated innovative approaches to conducting science and technology projects. These early OTA “wins” resulted in Congress enacting a series of programs, all of which required multi-party agreements and cost sharing, and expressly authorized the use of Other Transactions as funding instruments. Along with the programs came funding and a substantial increase in DARPA’s budget. The story of how DARPA handled this sudden increase in programs and funding by creating the Technology Reinvestment Project (TRP) will be told in Part 4.
***For more OTA information consider attending, The Need for Speed: Rapid Fielding using Other Transactions, September 26-27, 2018 in Arlington VA.