Costs Have Risen 200 percent

(In response to Kevin Galvin article in the Seattle Times: U.S. light-rail funding in '03 doubtful)

by Emory Bundy

As background, I'd like to provide a summary of the cost trajectory--which you characterize as 50 percent more than the original plan. It would be more accurate to say 200 percent more than the original plan.

I am using Sound Transit's numbers. Please bear with me:

1. The 14 mile portion for which we know the cost, the cheap, easy part, has more than tripled, from $900 million in 1996 to $2.9 billion today. Sound Transit publicly says its current, 14 mile MOS is $2.1 billion, or, in its February update of its 2002 financial plan, $2.2 billion. But that is simply what it calls its development, or capital project. It omits "other capital"--such as costs for acquiring the downtown tunnel, the promised Rainier Valley fund, public art, that sort of thing--which is $253 million. It omits project reserves, $138 million. And it omits debt service during construction, $351 million. That, according to Generally Accepted Accounting Principles, is part of the development cost of the project.

So the cost is not $2.1, or $2.2 billion, but $2.9 billion. You can verify it in appendix B of Sound Transit's New Starts submission to the FTA, October 2001 (where it's $2.84 billion), and in the agency's updated 2002 financial plan ($2.9 billion). Further, if you refer to the Inspector General's April 4, 2001 report, page one, and to the January 2001 report of Diversified Capital, Inc's review of the project., also page one, you will find the total cost at $4.16 billion. Sound Transit liked to say $3.6 billion, the difference being the same aforementioned items, other capital, project reserves, and debt service during construction. In doing so, it failed to include $560 million of the actual cost of the project. Long-term debt service? That's in addition, per GAAP, and begins when the project has been completed, and put into service.

2. The new 14 miles is not entirely the same as the previous 14 miles--though nearly six-sevenths of it is, 12 miles of the 14. The route has moved two-plus miles northward, at both ends. I don't know the cost of the northern addition--Lander Street South to Convention Place--but I doubt it's as high as the now-projected cost of the omitted two miles in the south, South 154th to South 200th Street, including Seattle Tacoma Airport. That cost is now projected by Sound Transit at $477 million.

3. The south 14 miles originally was supposed to cost about $900 million (1996). In January 2001, when it was reviewed prior to the FFGA signing, it was $1.56 billion. During 2001, it increased to $2.84 billion, and then $2.9 billion by February 2002. That's an additional $1.34 billion, just since the FFGA was signed in January 2001.

4. True, the maintenance facility is a costly item, and it now falls in the south 14 miles, whereas previously it was in the north MOS. I don't know its cost, but believe that $200 million is a reasonable, rough estimate. A shift of $200 million would still leave the recent, additional, projected cost overrun at $1.14 billion.

5. Why has Sound Transit's cost estimating been so extraordinarily wrong? The answers are found in a cost estimating performance audit it commissioned to its own, friendly auditor, Deloitte & Touche, which was completed in September 2001. Here's its summary of the problems:

… "Deficiencies in development of prior estimates including
… Development of estimates to match a budget
… Overly optimistic estimates
… Inadequate contingencies
… Contingencies prematurely reduced
… Inadequate/insufficient data (e.g. no soils data, ROW based on EIS, etc.)
… Inadequate soft costs."

By way of further explanation, D&T added: "the estimates were developed with too much optimism for the best case scenario to occur in all cases."

That explains why the south, easy segment was $900 million in 1996, nearly $1 billion in 2000, $1.56 billion in January 2001, and $2.9 billion in February 2002. Gradually, the cost estimating is getting more competent and honest. But there could be further shocks, as D&T think there's additional work to be done before the agency's cost estimating and cost control functions are adequate.

6. What are the implications for the rest of the project--the more difficult, more risky, more costly part? That has always been estimated at 60-plus percent of the total cost of the project. It includes extensive tunneling (while the south portion includes only 8/10ths of one mile of tunneling), and the bedeviling Lake Washington Ship Canal. The deficiencies Deloitte & Touche's audit identified, above, are systemic deficiencies, applicable to the entire system, not just the south segment. The southernmost two miles, in the SeaTac area, have skyrocketed, just as the 14 miles have. The only reason the five miles to the University District, the most difficult and problematic section, has not experienced similar cost increases is that no estimates are being made. Sound Transit hasn't even figured out a route. But the big news is, its quest for a less costly alternative alignment is coming up empty handed. The primary options now are the same alignment that was in the original plan, or a slight variation, tunneling under the ship canal somewhat further east, in the vicinity of the Montlake Cut. The advantage is, a shallower tunnel, hence the University station could be shallower, and somewhat less costly. The disadvantage is, it's a roundabout route--the Portage Bay tunnel is the most direct--and the added tunneling distance would add to the cost.

7. There's more. Sound Transit has accepted that it was a poor idea to run its trains so near the University of Washington's sensitive scientific equipment and laboratories, so a station appears likely to be moved from NE 45th and 15th NE, to Brooklyn, probably at NE 47th. And it has accepted that its plan to terminate its line right in the heart of the University District--a very congested, air-quality challenged area--was foolish. So now it is committed to continue at least to Ravenna Boulevard, which will require an additional mile of tunneling.

8. In the absence of data from Sound Transit, here's a reasonable estimate of the cost of the entire project:

Take the south 14 miles, which is budgeted at $2.9 billion, and assume it is now 40 percent of the cost of the total, 21-mile project. That's higher than it's previously been (37.5 percent), but that allows for the shift of the cost of the maintenance facility to the south segment.

Then the balance, Convention Place to U District, plus the two miles centering on the SeaTac Airport, 60 percent of the total cost, would be about $4.35 billion. This does not allow for the tunneling on to Ravenna Boulevard, which will be a substantial addition. But even at a conservative level, that suggests a total cost well beyond $7 billion.

9. Concerned citizens, including the Downtown Seattle Association, have pleaded for Sound Transit to complete its planning and cost estimating for the entire line, plus the requisite EIS, and have adequate funding in place, prior to starting construction. They have pleaded with the FTA not to proceed with its grant until the alignment is selected, reliable costs are estimated, and local funding is assured. DSA, in fact, thinks that the project should not proceed unless Sound Transit can assure--that is, it has good cost estimates and sufficient funding authority--it will get at least to Northgate, which will require something on the order of $1 billion more than anything mentioned in this memo.

Back to your article: It is vitally important that people here, and in Washington DC, have reliable information on this project, in order to make responsible decisions. Your suggestion that cost estimates have risen merely 50 percent or so, beyond that originally claimed, $2.3 billion, is very far off the mark. The cost submitted to the federal government in January 2001 was $4.16 billion--so the project was 80 percent over the original budget even then. Since that time, more than $1 billion has been added to the cost of the cheap and easy 14 miles, plus ballooning estimates for the two miles surrounding SeaTac. That pushes the project far past double the original price tag--entirely conceded in Sound Transit's data. And there's every reason to presume, confirmed by the Deloitte & Touche performance audit, that costs for the difficult north segment will similarly expand, so the biggest shoe has yet to drop.