How Much Risk Can Your Vendor Take?
- By Arielle Emmett
- Nov 18, 1996
In state information technology circles, performance-based contracting is hot. In most arrangements, vendors front the bulk of the money to build an information system and agree to paybacks only after the systems perform above par. But now, with a little experience under their belts, both vendors and public IT managers are wondering how to divvy up risk more evenly so that everyone benefits.
Consider California's Business Entities Tax System, a joint project of the state's Franchise Tax Board and Andersen Consulting Inc. The system, designed to cross-reference and validate state business tax returns, would generate new revenue streams through more accurate accounting. In its contract with the tax board, Andersen took virtually all the up-front financial risk. "If the system did not produce the intended benefits, [Andersen] would not get paid," said P.K. Agarwal, the board's chief information officer.To set up the deal, the tax board created what it calls a solicitation for conceptual proposals.
While it sounds more like a think-tank grant application, the SCP is actually a request for proposals with a much less detailed spec. "We were more interested in outcomes than specifications," Agarwal said. "What we were offering was a fairly flexible way for vendors to be creative in how best to solve our problem. This is a risk-based partnership; depending on the nature of an outcome, [the business partner] gets paid accordingly."
The mechanics of the contract are this: Andersen invested $22 million - the contractual amount - in hardware, software and services. Out of the revenue generated by the system, which far exceeds $22 million, Andersen will be paid up to $22 million - plus interest on the outstanding balance - based upon whether the tax board realizes expected benefits during a five-year period.
"If the benefits come in faster, we get paid faster," said Ken Dineen, an associate partner with Andersen Consulting, Sacramento, Calif. "The arrangement allows for the fact that you can go a full five years, and if the government doesn't get full benefits [from the system], we would not receive the full amount of the contract." That is unlikely in this case. "The system is producing the kinds of revenues we expected it to," Agarwal said.
Such arrangements force business partners to solve the whole business problem, not just provide a technological solution. "Traditional [government IT] contracts have not been optimizing the value of the systems or the results," Dineen said. "Part of the reason behind this is that both parties weren't incented equally to perform."
The new - and riskier - performance-based plan forces vendors and integrators to work with agencies more in the planning phases. "It allowed some of the best thinking from both the client and vendor as to what's the best solution," Dineen said.
Not every vendor is willing to take such risks, however. Many prefer to invest a comparatively "safe" amount - $1 million or so - in smaller-scale "demonstration projects," with only modest payback incentives. An example is Electronic Data Systems Corp.'s red light automated ticketing project in San Francisco, which calls for automated cameras to capture motorists running red lights. Such projects provide proof-of-concept work and revenue from ticket generation for the vendor. But they don't necessarily provide the guarantee of a larger contract.
"On the surface, in its simplest form, performance-based [contracting] sounds like a wonderful thing to do," said Merv Forney, president of EDS' state and local government section. "The real pebbles are in the details."
Management influence in the partnership is also a success factor. "If I'm to spend my money on a project to make someone successful, I must not only provide the expertise but have a partnership say in the way an organization is run," Forney said. "I can do all kinds of re-engineering, but if I don't have some say or control, then the whole risk-reward structure will fail. You can't go at-risk financially on the success of an IT project without some say or authority in the decision-making and in the running of the enterprise."
EDS' strategy thus far is to limit up-front investment to performance-based systems they can readily control. For example, the company has set up incentive-based alarm-management systems for police departments in Houston and Charlotte, N.C., at a cost of $1 million each. Designed to reduce the number of false alarms, help track data and issue citations, the systems are implemented with EDS' money. "We get paid out of the collections of the citations," Forney said.
Although that can be lucrative, some experiments are not. In the case of the San Francisco automated red light ticketing project, which is intended to last only six months to a year, "we haven't made any money yet," Forney said.
But even some money-losing projects can be good investments; they can be justified to set metrics for a new technology. They also can "give governments an opportunity to filter through companies that can perform in a credible fashion vs. those that cannot," said Sharon O'Malley, EDS' vice president of sales and marketing. "It's the 'fly before you buy' [approach]," O'Malley said. "The advantage for private industry is that you're showing your commitment to a solution."
Nonetheless, some vendors believe they have been burned by too many public-sector ventures. "A lot of money has been spent on projects that never came to completion, cost too much and weren't done on time," Forney said. "You've heard all kinds of war stories with millions spent and nothing to show for it.
"Usually it's a mismatch of customer expectation vs. what vendors think. Customers are asking for 'x,' and vendors are hearing 'y'; a lot of times with contracting, our experience is that for a fixed price, [customers] want everything. Every time a change is asked for, they expect the private sector to do it whether it was in the original specification or not."
As state governments reach for more innovative IT financing solutions, competition for private-sector support may grow tougher. Vendors may want higher performance incentives to take greater risks. At the same time, private investment firms and venture capital companies may be a readier source of money than suppliers or integrators.
"The private sector can only enter into these risk-reward relationships when there's a high reward on the risk in a relatively reasonable period of time," Forney said. "I tell my customers they can go borrow from banks cheaper than they can borrow from me. [Our question is], how do we invest our corporate money into projects that have the highest rate of return?"
How indeed. EDS soon will be grappling with these questions in Texas, where vendors are teaming with state agencies to overhaul the state's welfare-delivery system (see story). "The uniqueness of this is that the private sector is not only teaming with public-sector agencies, but ultimately the public-sector agencies will be competing against each other to manage a new type of welfare process, which focuses on getting people back to work," Forney said. "What performance incentives there may be will be incentives based on the amount of dollars saved and people put in jobs. Even before the contract is awarded, we're spending money. [We see the contract] stipulating a combination of basic services and being paid for performance goals."
Public-sector chief information officers regard performance-based contracting as "a fairly interesting new ball game," according to California's Agarwal. In some cases, he said, they are finding that "the more expensive solution may have a better return on investment." So up-front specifications and rigid rules of contracting are giving way to more creative problem solving.
Penalties for Success
Ironically, the success of private financing might, in some cases, create pressure to undo such arrangements, experts warn. "The negatives are that the public sector is very concerned about control and about the appearance of impropriety or any favorite treatment," said Jerry Mechling, director of the Program on Strategic Computing and Telecommunications in the Public Sector at Harvard's John F. Kennedy School of Government. "These things are too risky for the government to do on its own. Yet if the private sector accepts the risk and makes money off it, government often comes under pressure to renegotiate the agreement and take the money away.
"The public sector and its oversight process is uncomfortable with paying for risk," Mechling continued. "However, the good news is that partnerships are potentially a way to pay for risk. And the government cannot afford in today's environment to simply allocate money and take the risk itself."
Expect more such experiments to take place in the near future. "I think it remains to be seen how big and how long and how well this will develop, but it's clearly developed in the private sector," Mechling said. "Very large systems efforts in the private sector have often evolved to be partnership arrangements, [where] one firm knows the systems-building and the other knows the business for which the system is being built. Instead of having one firm assume all the risk and just buy what's needed from the other firm and own the result, a number of cases are going forward where they partner together, and both will own the result.
"In the public sector today, lots of initiatives have been taken in that way, but it's very new. I think, frankly, [that] while it's attractive, the dance remains to be played out."
Arielle Emmett is a Wallingford, Pa.-based free-lance writer who specializes in computer and network technology. She can be reached via the Internet at firstname.lastname@example.org