Organizations are like people. When in an impossible situation, they may respond in irrational and unreal ways. Such is now the case with the California State University system ("CSU"), which is on the verge of creating a joint venture with four major corporations to re-wire the University system for 21st century tele-communications. The venture, called "CETI", is to be led by GTE, a major telephone and wireless enterprise. Fujitsu, Hughes Satellite and Microsoft fill out the ranks. the
The education bureaucrats who run the CSU from the Chancellor's office in Long Beach have recognized the importance of the Internet for several years. They are also acutely aware how averse California politicians and voters are to paying taxes to support public higher education. The estimates vary widely, but the current plan projects a three year need for $300 million in computer and telecommunications network upgrades. Meanwhile, the faculty restlessly seek to recoup their losses from the pay freezes of the early 1990's, the buildings are slowly crumbling, and the legislature keeps paring away at the budget requests. Under such circumstances, it is only natural that the CSU managers would turn to off-the-books financing, and try to convince themselves that partnering with Corporate America offers a magical solution to the problem of "no money".
Nor is it surprising a company like GTE, seeking to make its way in a newly competitive telecommunications business, would express interest in such a partnership. An executive who does not appreciate the diverse nature of the CSU's 23 campuses might dream of 320,000 students converted into new customers with a stroke of the pen. The prospect of inter-campus communications links centered in the Long Beach headquarters, of a centralized system-wide Help Desk, of a self-funding entity allied with the CSU yet immune from the legislative budget committees - - these prospects must be powerful visions in Long Beach. This despite recent moves to decentralize, which recognized the reality that the size of the system makes centralized management unwieldy.
In September, the CSU chose the "GTE team", as partners in the CETI consortium. The details are to be worked out in marathon bargaining sessions from now to December 15, 1997. The best guide to the project as it now stands is the GTE sponsored business plan, available at a San Jose State web site. A careful review of that plan reveals fundamental flaws in its core assumptions. It is time someone stood up and said "But he has nothing on."
It will take a lot of money - - about $300 million over three years - - to build out the telecom/Internet connections across and between all 24 CSU campuses. Part of the expense may be because the proponents chant the mantra of "everywhere, any time service". This is a good sound bite, but may be bad facilities economics. It might be that the cost of 99% coverage is far higher than 95% coverage, as the remote outposts are the most expensive to bring into any system. There is no explicit analysis whether the marginal cost is worth it.
Financial analysis is a surprisingly simple exercise. Almost anyone can play. The only question is: "Where does the money come from and how will it be paid back?" Chapter 16 of the business plan contemplates three sources. First, the corporate partners will put in $36 million for "working capital", to cover the time lag between expenses and revenues. Next, "a company such as GTE" (how modest!) will build part of the improvements and then lease this equipment, cabling and wires to CETI -- or maybe to CSU itself (16-2). The lease might be direct or through an intermediary investment bank. The plan assumes "a simple lease with a conservative 9% lease rate" on a $120 million investment, which just means that someone (CETI with CSU backing, or CSU itself) would be obliged to pay $10.8 million every year to GTE or to its banker. (It is not clear whether the $120 million is the cost of the system or is what GTE wishes to be paid - not necessarily the same thing.)
The plan is silent about what happens when the ten years are up. The equipment and cables will no longer be the state of the art: indeed, Table 16-4 on page 16-7 shows this infrastructure will be fully depreciated by the year 2007. (A modest slice of anticipated profits are to be set aside for "system refresh".) This system will be an essential part of day to day education and administration at the CSU. However, GTE or the banker will still own the system, and will need to be paid for it. The $10.8 million per year appears to be only a financing cost, like interest. No rational investment banker would put up $120 million dollars for 10 payments of $10.8 million - that's a clear loser even if we don't count the time value of money. At a rate as low as 7%, it would take over 22 years of $10.8 million payments to recover a $120 million investment.
At some point, then, the principal as well as the interest will have to be paid. The idea that the CSU could stop paying is ludicrous. Will all the electronic classrooms turn dark? Will we give up telephones and computers and communicate with signal flags? Will the repo men swarm over the campuses, pulling up copper wire and fiber optics to sell for scrap? As a practical matter, the CSU will be on the hook for the lease payments, whether interest or principal. If CETI revenues fall short, the CSU must make up the difference, or shut down.
This is a "special purpose" capital lease. Financial accounting long ago recognized that this type of lease is functionally identical with a purchase of equipment which is paid off over time. FASB Statement 13 requires that the equipment be carried on the books as an asset and that the payment obligation, as calculated from the lease terms, be shown as a type of debt. One searches in vain through Chapter 16 and the Appendices to the business plan for any recognition of the $120 million long term liability associated with this financing.
The final financing component of the plan is to float a private placement of $180 million in ten year bonds with a 7% interest rate (16-2). Since CETI is a novel blend of the state education system and profitseeking enterprises, there are grave doubts whether this borrowing could enjoy the Federal income tax exemption which now allows the CSU and the State of California to borrow at favorable rates. Another $12.6 million flows out each year for interest, with a balloon payment of all the principal coming due in ten years.
How would all this be paid for? The rules of this peculiar game do not allow CSU to show on the books that it is obliged to pay $23.4 million in annual interest or the lease equivalent on $300 million of financing that will some day come due. Instead, the business plan proposes three general sources of revenue:
The money must come from the other two sources. What on earth, one may ask, does "CSU contractual support for which CETI will be tasked to perform" mean? Table 16-2 shows that CETI would absorb all telephone, computer support and communications functions throughout the CSU, except for the main frame data centers. It appears that CSU headquarters surveyed the 24 campuses and was told that all telephone, computer support and communications functions cost $122 million per year: $27 million went for central telecom, $9 million for off-campus access, $16 million per year for user training, $15 million for "instructional media", $32 million for "schools and colleges", and $21 million was "other". It appears that the numbers are a bit spongy and the survey methodology imprecise. Nonetheless, it seems that the financial success of the plan hinges on running all campus Audio-Visual operations and all School computer labs from CETI's headquarters in Southern California - no doubt, in the vicinity of Long Beach.
The GTE plan, perhaps not accustomed to academic imprecision, treats the $122 million as a line item hard-dollar budget number. The plan then pulls some cost reduction factors, seemingly out of mid-air. By the year 2000, the cost of user training and school/college support is to be reduced 35%, with 30% cuts in telecom and instructional media spending. Table 11-5, page 11-3.
Workers need not fear, however. You are repeatedly assured that these cost reductions will happen without anyone being laid off. One wonders a bit, however, because the "People Plan" (Chapter 14) keeps taking about "human resource assets". Perhaps our colleagues in Psychology can diagnose the syndrome behind this weirdly depersonalized way of describing the faculty and the staff who work at the CSU campuses. Alternatively, perhaps the notion of introducing "flexibility into the mobility of their human resource assets" (Sec. 14-4) is simply technobabble for good old-fashioned union bashing. Students, too, are to be "utilized to fill personnel and functional shortfalls", and are to be given degree credit for "participation in working for CETI". (Sec. 14-6)
So how does CETI make any money from taking over telecom and computers within CSU? The answer is: by keeping all cost savings for itself. A wave of the wand is to cut $30 million per year from the cost of providing these services, fully effective by 2001. CETI, however, will continue to charge CSU the old price: the telecom service contract will bring in $122 million each year and only cost $83 million. The contract profit pays for CETI overhead, debt and lease service, and helps the buildup of $106 million of retained earnings by 2007 on the initial $36 million investment. (Table 16-5)
With no layoffs, how is this enormous cost savings to be accomplished? Chapter 11 discusses such measures as increased physical plant utilization (translation: running the University during summer), "administrative systems enhancements and related enhanced capabilities", and the reduction of redundant telecom programs (Table 11-1). It appears that CETI will appropriate for itself many efficiency moves proposed at various times by CSU consultants, which have little to do with telecom systems (page 11-6). For example, Student Services costs are supposed to go down 20% as paperwork processing is reformed and duplicate signatures or reviews are eliminated - based on a Coopers & Lybrand efficiency study of applications, financial aid and registration.
But if CSU were to re-engineer its paperwork by itself, it would keep what it saved. Why turn the money over to CETI? Now comes the kicker - the people doing the work will be CSU employees under subcontract to CETI. The new entity will not have any employees of its own, except perhaps some executives who will be granted the obligatory stock options (Sec. 12.5). CETI will also have rent free access to CSU facilities, including power, heat, lights, services and security (15-1). Absent employees, the work to be done will be subcontracted to CSU. For example, the computer labs will belong to CETI, while the lab staff will be CSU employees working for the CETI Technology Services Utility (Fig. 12-2) under a subcontract, so CETI can deliver the promised services back to the faculty and students of CSU.
Suppose you want a fence painted. You hire your neighbor to paint it and agree to pay your neighbor $100. Your neighbor then hires you as a "subcontractor" to paint your own fence for $70.
Nice work if you can get it.
