At recently as 2012, Sound Transit had less than one billion dollars of debt. That increased to $4.6 billion by the end of 2017. The ST3 program will push it much higher, peaking at a projected $17.6 billion in 2035. That path puts Sound Transit close to legal limits as major ST3 projects are delivered. After 2035, as most major projects are completed, the outstanding debt could be reduced.
Sound Transit’s ability to take on more debt is constrained both by statute and by policy. Most immediately relevant is a statutory limit of non-voted debt at 1.5% of the assessed value of property within the RTA. This is the same limit all Washington State municipalities face. With 60% voter approval, the statutory limit would increase to 5%, though other policies and bond covenants would prevent the debt ever getting that high. The cushion between actual and allowed debt is most narrow in 2035, when the $17.6 billion in outstanding debt nudges against the $20.1 billion limit.
All financial projections over such an extended period are uncertain, combining assumptions about revenue growth, federal grants, project costs, and interest rates many years into the future. The legal debt limit may be lower than anticipated if the growth of assessed property values slows. Some risks correlate in unhelpful ways. Prolonged slower growth would mean lower tax revenues and less ability to borrow to fill the gap.
The long-term debt plan is very sensitive to small changes in the financial plan in the early years. Consider the bills to correct MVET valuations. The tax revenue loss from updating the valuation schedule is just $780 million between 2017 and 2028 if there is no other mitigation from the Legislature. The implications for Sound Transit debt are larger. Absent offsetting cost reductions, the lost revenues must be replaced by debt. The average interest rate is anticipated at 4% through 2021 (reflecting low recent bond rates), and 5.3% thereafter (a more typical pre-recession cost of debt). Sound Transit pays a 1% origination fee on bonds and typically begins repaying principal after five years. The principal and interest payments are themselves funded with more borrowing. At those rates, the MVET reduction accumulates to $1.6 billion in outstanding bonds by 2035, or $2.2 billion by 2041. That appears to just fit within future debt limits, but the margin of error is much too narrow for comfort in a decades-long financial forecast.
Could Sound Transit borrow to replace a loss of federal funding? Lynnwood Link and Federal Way Link alone were anticipated to receive $1.67 billion of federal grants, with several billions more in future New Starts loans on other projects. There is some short-term flexibility in the plan, as Sound Transit is still far from its maximum debt capacity. But extend out the cost of replacing those grants with accumulated interest, and the debt impact roughly doubles by 2035. More debt in the 2020s means exceeding debt limits in the 2030s. A significant shortfall in federal funding would almost surely mean delays or scope reductions somewhere in the capital program.
Not all risks are downside. Current growth is strong, though the upside to tax revenue is partly offset by construction cost inflation. State legislation enforcing sales tax on internet sales may yield tens of millions of dollars per year not in the current plan.
Transit advocates should watch closely for cost overruns and scope creep. If costs are higher than the representative projects assumed in the financial plan, the Board would restore financial health by cutting or delaying other projects.
In 2015 and 2016, the Sound Transit Board was challenged to meet a long list of desired projects. The response was to stretch the program, shoehorning in as many projects as possible, and then to deliver as quickly as finances allow. Just between March and June of 2016, the plan grew from a $50 billion draft plan to $54 billion without additional funding other than more borrowing. The high debt load leaves little margin for error in delivery. Keeping the program on schedule will require careful cost discipline as well as favorable outcomes in the other Washington.
Hence, the point of “design-build”. If the contract with the contractor calls for recurring payments, rather than a “lump sum”, the ongoing “operating cost” money can be padded to include what is effectively interest on a contractor-issued loan to complete the project. But, since no money is officially borrowed, it doesn’t count as “debt”, and doesn’t count against the statutory debt limit.
Maybe my problem with “design build” is based on a mistaken understanding. But to me, term suggests that a transit system is handing over decisions that are rightly its own responsibility. If you want anything done right, like a major decision….remind me of general best choice of a contractor?
But meantime, for general shape and direction of the work, what the ST Board doesn’t know will still leave survivors after they learn it. At least enough not to leave the dead wire caused by a substation breaker, that drivers can’t power under, half a block up the Queen Anne Counterbalance.
Or at least not to leave it there for forty years.
Mark
What kinds of contingencies were built into ST3 cost assumptions? That affects the viability of the program. It appears that Lynnwood and Federal away had contingencies too low — and if the other projects have that issue, there’s definitely a big problem.
However, the 2021 reapportionment will create a major shift to the ST district in the legislature — more sympathetic to transit. I think the level of cooperation in Olympia to get through cash flow issues will be better. Since the same reappointionmrnt thing is looming to a lesser extent, DC may also be more sympathetic too.
The 2017 Annual Financial Plan used the following assumptions:
>>>6.2 Summary of Financial Assumptions
•Sales Tax Rate: 0.4% 1997 – 2009; 0.9% 2009 – 2017; 1.4% 2017 – 2041 (subject to potential sales tax rollback, after 2041)
•Sales Tax Base Average Annual Growth: 3.8% 2017 – 2041
•MVET Tax Rate: 0.3% 1997 – 2028 (tax ends in 2029)
•MVET Tax Rate 0.8% 2017 – 2041 (updated depreciation schedule in 2029)
•MVET Tax Base Average Annual Growth: 1.9% 2017 – 2041
•Rental Car Tax Rate: 0.8%
•Rental Car Tax Base Average Annual Growth Rate: 2.9% 2017-2041
•ST3 Property Tax Rate: $0.25 per $1,000 of assessed value
•Average fare prices increase at a rate generally in line with inflation
•Sound Transit to receive $7.7 billion in grant revenues 2017 – 2041
•CPI Average Annual Cost Inflation: 2.3% 2017 – 2041
•CCI Average Annual Cost Inflation: 3.4% 2017 – 2041
•ROWI Average Annual Cost Inflation: 4.8% 2017 – 2041
•Interest Rate: 2% on General Fund; 2.5% on Reserve Fund Balances
•Bond Interest Rate: 4% 2017-2021; 5.3% 2022 – 2041
•Bond Term: 30 years
•Interest Only Payment Period on Bonds: First 5 years
•All-in Issuance Costs: 1.25% of Par Value
•Principal Set-Aside for Bond Reserves: 7.15%
•O&M Reserves equal to two months’ O&M costs<<<
Why does ST go five years interest-only?
That’s a common financing strategy used with serial bonds.
“What kinds of contingencies were built into ST3 cost assumptions?”
Explicit contingencies [in the ST3 program financials] are 7%. Some observers will tell you there are hidden contingencies in padded cost estimates. That’s not easy to evaluate unless you’re confident you know better than ST Planning what future costs will be.
FTA recommends 30 percent at this point. Even with padding, 7 percent is way too low for such an early stage.
I think that it’s fairly certain that we are going to have to face the fact that we can’t get what ST3 promised without unanticipated Federal or state money — at least $5-10B more.
FTA recommends since when? Is this a longstanding recommendation, or a recommendation due to the changing landscape the past year, which appeared too late for ST3’s budget to consider, and now ST has limited maneuvering room within its overall budget constraints.
Page 11 of this September 2015 FTA document has a nifty graph explaining contingencies.
https://www.transit.dot.gov/sites/fta.dot.gov/files/docs/OP40c%20Risk%20and%20Contingency%20Review%20%28Full%29%20-%20Sept%202015_0.pdf
Since we aren’t even into engineering, we’re pretty much at the start of this graph, which is 30 percent. Here’s a quote from the same document:
“The FTA has determined, from historic project information, that the following minimum levels of contingency (the aggregate of allocated and unallocated cost contingency) are prudent:
• At Entry into Engineering, 25% ”
I think it’s important to note that major transit construction has many, many things that cannot be reasonably estimated at an early stage. It’s not like other specific grant programs like for buses or signals, which do not have lots of guesses about where and how things will go and who will need some sort of mitigation.
Sound Transit issues many of its contracts using a 10% contingency standard. See Motion 2017-119 for an example.
Project planning timelines are based on the average of ST1/2 experience. Construction timelines have a year’s padding for complications. I don’t know about budget contingencies, but losing all federal grants and having the state yank promised tax streams goes beyond reasonable contingency planning. ST could have made a budget assuming no grants, but ever since the beginning it has declared that it will leverage federal grants. If it didn’t, ST3 would have to be smaller by a project or two, and the public was begging ST to get many projects in because they were all high-priority and urgently needed. It was also assumed until election night that Trump would never win and that Congress wouldn’t agree to eliminate grants because it has always failed to do so. (Thank you “never Hillary” moderates who voted for Trump.) But the public knew that a major recession or losing grants might happen (because the grants might be awarded to another city), and that that would cause deferring projects or extending the timeline, as happened in the 2008 crash.
South Kirkland to Issaquah is not “high-priority and urgently needed”. IF Bellevue creates big nodes at the three stations along I-90 and IF Issaquah continues to densify, having the line will have been quite valuable (and of course will still be) in 2050. Absent it, those big nodes wouldn’t have been created.
But not to put to fine a point on it, but there is, at this time, no direct bus from Issaquah to Bellevue. Metro hasn’t seen fit to establish the express that should “pre-shadow” the route.
If for no other reason than to see if anybody wants to ride. Jes’ sayin’.
I’m yet to see any reason to believe that we’ll lose all federal grants. I understand the administration has zeroed them out, but I also understand that Congress controls spending in our system, and they have shown little interest in following along. Courts won’t allow the FTA to apply irrelevant criteria to grants, such as zeroing out grants for ‘sanctuary cities.’ No grants is a worst-case scenario that is unlikely to come to fruition.
Richard Bullington
ST 556 is direct from Issaquah to Bellevue TC, and the 271 Not an Express but all day service
The Issaquah line is an extra project at the end, like Tacoma 19th Avenue. ST scheduled them last, presumably because it considered them the lowest priority and the first to be cut or rolled into ST4. Putting these projects up as the reason for ST3 or the reason people voted for ST3 is the tail wagging the dog. ST3 exists and people voted for it mainly because of Everett, Tacoma, Ballard, West Seattle, 405 BRT, and 522 BRT. Those are high priorities, at least according to ST and the people who give feedback to ST.
Richard, your position is an ideal one. I agree with it theoretically but it’s not something this region’s governments or public are willing to do at this time. The Totem Lake and Issaquah regional growth centers are being driven by cities that don’t want to be left behind in economic prosperity (all those future companies and residents paying taxes to the local municipality), and in Kirkland’s case to keep growth away from downtown and privileged areas. Those are far larger and longer-term issues than we can discuss in an ST3-approval-and-initial-plannig timeline. Hopefully in the 2020s or 2030s the public mood will change and we can revisit some of these decisions as a new generation comes to power.
I can understand why Richard didn’t realize that ST has a bus connecting Issaquah to Bellevue, because he probably has never met anyone who has ever made that trip. I get it. You have a better chance meeting a member of the Eritrean Winter Olympic squad.
OK, it isn’t that bad. The 555 has 62 people a day making the trek from the 6 Eastgate/Issaquah stops. The 556 has 315. So, basically less than 400 people a day ride the bus(es) that will someday be converted to a train.
Oh, and if your bus number is bigger than the number of people who ride it, maybe it isn’t a great bus and maybe you shouldn’t worry about replacing the bus with a train.
Poor school bus routes in the 6xx or 7xx!
The 271 goes to Issaquah? Well, then, I am certainly wrong. How many people ride “across” Bellevue College?
The 556 doesn’t count; it’s peak only and a rail line, unless it’s commuter trains running on freight track, can’t subsist on peak only.
The 271 does go to Issaquah, although it takes 45 minutes from Bellevue TC and 73 minutes from the U-District. Earlier it was truncated at Bellevue TC part time (Sunday and evening?), but now only runs above 30-minute frequency are truncated.
The last time I went to the Issaquah Highlands to see how the new urbanist village is doing, I decided while returning to go to Bellevue instead of Seattle. I took the 554 westbound to transfer to the 271 at Issaquah TC, but it would have been a 20-minute wait and a 45 minute bus ride, so instead I went to Mercer Island and came back on the 550, which runs every 15 minutes midday and doesn’t take as long.
@Mike — Ha, yeah, school buses don’t count :)
@Richard — Yes, the 555 and 556 are peak only, but the point is, they perform poorly, even for peak buses. They never come close to being full, even during peak. They average less than 20 people per bus ride from the Issaquah/Eastgate corridor. That is very bad for peak, and if the bus ran all day, it would be much, much worse. The 535 runs from Lynnwood to Bellevue, and carries about 40 people during peak, but dips down into the teens the rest of the day. The 532 (Everett to Bellevue) is rush hour only, and gets over 70 people per bus. That means there are lots of people taking the buses from Snohomish County to Bellevue during rush hour (between 40 and 70 per bus) but very few the rest of the time (in the teens).
You just can’t find buses that hold steady across the board — they all have big peaks during rush hour. The ones serving the suburbs usually have the biggest differences. It is quite likely that if they ran the 555 or 556 all day it would get ridership in the single digits.
Poor services gets poor rider ship. Issaquah has committed to density and has been growing at 40% per decade for the last 3 decades. If trends continue by the time Light rail gets to Issaqauh their could be 100,000 people living in its 11 sq miles which would more then qualify it as dense enough. The demand will only grow as Issaquah becomes a commuter community to Bellevue. Remember ST planing Light Rail for the future not just the neglect for the past.
The Issaquah to S Kirkland line is dubious at best.
It might do slightly better than current buses if:
1. Riders to/from downtown Seattle can be convinced to take East Link and transfer to the new Eastside rail line.
2. Commuters to Microsoft can be convinced to take the train (with a transfer)
3. There is some latent demand that frequent rail service will induce
4. Congestion gets really bad and parking very expensive in the areas of the Eastside served by rail.
Honestly this line is only happening because there needed to be a major ST3 project for the East Side.
What needs to happen is scrapping the plans for this and 405 BRT. Go back to the drawing board and come up with a world class BRT network for the East Sub area.
[ot]
Over 17 years good or ill, nobody can truly know. So best we can do is run the project as conservatively (New Electric Railway Journal for “at least check out light rail before we go heavy”) as we can. But above all, like any good athlete or worker, eyes open, reflexes sharp, ready to flex without losing momentum or direction.
AlaskaUSA Lending Department will be waving the next loan approval for my 90th birthday present.
Mark Dublin
A country is not a corporation: it has responsibilities beyond having a balanced budget. The Federal Reserve exists to ensure the country can meet them, as well as managing inflation and unemployment. Good debt is infrastructure and human investments: things that will make us more successful and resilient in the future. Bad debt is tax cuts so that corporations can buy back their stock and CEOs can become oligarchs. But one term of bad debt is nothing to panic about, if there’s a reasonable chance it will be significantly reversed in four or eight years. We’ve now had thirty-seven years of Republicans running up the debt for tax cuts and wars and Democrats in between bringing it down (while also doing something for transit and human capital, even if minimal). So there’s reason to believe it will probably happen again.
Thanks, Dan. Would really be great if WordPress can develop a red pencil line through a word, though. All I meant was to stress that in a lot of ways, what you are is as important to success as what you do.
Mike, you and I are completely on same page about a balance sheet being same as a Bible. Easier to thump than read. Also, though, many years trying to figure out long winning streak of the Republicans. And much worse, the Southern Democrats who killed them and stole their party.
Having watched our politics swing steadily rightward my whole life, though, really am seeing the importance of a single idea, repeated long enough that it goes from being something to think about, to something to think with.
Best lifetime example: I turned voting age at a time when many people believed that Government was neither a tyrant nor a benefactor but rather ours, the people’s tool for accomplishing our purposes. Key word: OURS.
Now it’s a THEM for the rich to protect us from. Which thought has the unbelievably powerful pull of relief from responsibility. Think experience shows that it’ll become OURS again when it no longer fits anybody else’s business plan. Shouldn’t be long.
Mark
As Thomas Picketyu said, the two world wars and the Depression broke down the hereditary oligarchy, and when people returned form WWII having worked on equality with others on a common task (and government-funded), they took that attitude into civilian life and started eirking together for the good of everybody (“Out” society). They had a blond spot for minorities but that doesn’t negate the rest of it. That system weakened in the 1970s and gave way to, um, right-wing rattling, and the result was a new round of oligarchs, who in a generation or two will become hereditary again. And they want you to believe they’re one of “us” rather than “them’.
The City will need to step up, especially if Democrats lose the 2018 elections and then Trump is re-elected. Federal investment in “blue” cities will disappear.
The City has the right to issue a billion in bonds for “fixed guideway transit” which is not light rail, however it’s defined in the statute. And it certainly has the financial capacity to amortize them. Could such bonds be used to dig the new tunnel since it could certainly be converted to heavy rail just as the bus tunnel was converted to light rail?
Trying to think outside the box.
It can raise $1 billion I’ve heard. `You can decide what you want to spend it on.
The grants aren’t gone yet and there’s no certainty they will be. So we shouldn’t assume now that they’ll be gone, we should just do some worst-case scenario planning.
I want to see if I’m understanding this correctly. It looks like ST’s bonding capacity this year is right around $10B. So the total valuation of all ‘real’ property in the ST district is ~$666B? Which would be around 238k per capita (population of 2.8MM according to https://www.soundtransit.org/blog/platform/q-what-sound-transit-district.) And that total valuation should double by 2035?
Assuming I’m getting that correct, what’s the source for that line on the graph?
I assuming it is the 1% that property values are allowed to increase year over year
From the ST Financial Model, the assessed value in the RTA for 2018 is $637 billion (for a $9.55 billion bonding capacity). ST assumes a 4.49% growth rate in assessed value going forward, which gets to $1,343 billion in 2035.
Did they just make up 4.49% growth, or is that based on an analysis of population growth and historic property value increases? I’d like to understand the assumptions that are going into that projection.
I’d assume there’s an analysis of history, but the documents I have don’t specify what they considered. The assessor’s offices would have the historical data for the Counties or the RTA.
Ok, oh well. Thanks for answering, though.
It isn’t clear to me what happens if there is a significant loss of revenue. It is fairly easy to see what will happen to the spine. It will be shortened, as has been the case in the past (when South Sound revenue was less than expected). But how do you shorten the Issaquah to Kirkland run? Maybe go from Kirkland to Eastgate. That would cover the key parts (and arguably be better overall for riders than the full thing), but would still cost a fair amount of money. I am guessing you might save about half the money that way.
But what about Seattle? Maybe run it from Ballard to SoDo? That is by far the best value, but still very expensive.
In both cases, how much leeway does the ST board have in building something completely different? Putting politics aside, could ST build Ballard to UW instead?
I’d bet that it wouldn’t be Ballard to SoDo that gets built but Smith Cove to some connections south of IDS. That’s the critical section that needs more capacity. In extremis the City can start enforcing the Red Lanes.
To help out Ballard, I should have said. .
I see what you are saying — build the new tunnel and run the train on the surface. In that case, you might as well go to Interbay (basically everything but the new bridge). I don’t think you save much by going to Smith Cove, and there is really nothing there. There isn’t much more at Interbay, but at least you have good connections to Magnolia, SPU (and Fremont, via the 31/32 or something similar).
But that really is my point. Build that and you haven’t added that much for your money. Sure, Lower Queen Anne and the two “South Lake Union” stops are fine, but this is a spur line, after all, that might require a very time consuming transfer. For a lot of trips, the surface buses will be better. It is just a lot of money to spend on something like that, as opposed to Ballard to UW, which would improve an entire region (everything north of the ship canal and east of I-5).
I think that ST has made a commitment to get to Smith Cove and serve Expedia (or whatever is in that complex in a decade). The idea is to avoid the Ship Canal crossing if there isn’t enough money to tunnel it. Better to build less better than something that people will regret for a hundred years.
I think people are making too much of the drawbridge. A 70′ bridge isn’t going to open that often.
I’m guessing most likely the timelines will be extended for both West Seattle and Ballard. Scope reduction will likely come in the form of “value engineering” which means no tunnel under the ship canal.
The other option is sub-area borrowing since Issaquah-Kirkland is a non-priority (as are some of the Pierce County Projects). This may allow Ballard/Downtown/West Seattle to proceed on-time.
MAX ran mostly just fine at 7 minute headways with a single track line and passing sections of double track east of Ruby Junction. With a wood chip dock at Lake Oswego, a paper mill in West Linn and several other industries sending large barges of freight through the Steel Bridge every day, eventually they had to change it.
I see no reason why Issaquah to Kirkland needs to be double track the whole way. It has no drawbridge to throw off the schedule and probably won’t have the density to justify anything more than 7 minute headways or so.
Part of the reason to go to Ballard is to get transit separated from the congestion on the Ballard Bridge. They might wind up with Ballard as a ground level median line on 15th though. Could be the fate of West Seattle too.
I’ve often wondered if Issaquah could be served by a mostly single track line and slightly less frequencies. It could financing that would enable the line to open years earlier.
To do that would probably require a third platform in Bellevue. I don’t see anyone talking about a third transfer platform option at one of the three joint Bellevue stations (not even a switch!) to allow for a mostly single-track operation that could work without compromising East Link train arrivals. So we’re left without a timed transfer point for Issaquah trains and that makes single-tracking much less logical.
It’s tragic because we could get this facilitated with some East Lonk design modifications before the line opens. Instead we have to pay more and experience service disruptions years later.
Eventually it will probably be extended to downtown Kirkland, Totem Lake, and maybe Bothell. At what point would a single track not be enough?
Having the Ballard line run on the surface is not bad at all. There aren’t that many significant cross streets (unlike MLK Way). But that really wouldn’t save that much money (a couple hundred million if I remember right). The expensive part is the bridge and the tunnel. Likewise, there is no cheap way to run a train line out to West Seattle. You still need the tunnel, and a bridge of some sort. Running out Delridge is cheaper, but that is significantly different than what was originally proposed. At that point, you might as well build Ballard to UW (since that too is significantly different than what was proposed).
Getting rid of tunnel sections is the most major way to keep ST3 corridors intact if big cost reductions are required.
One crazy cost-saving, politically-difficult idea would be to replace the monorail tracks with light rail tracks between Westlake and Seattle Center with intermediate stations at Denny and at Bell St, Relocate the monorail system somewhere else, and engineer a connection to Smith Cove and to a portal across Pine/Pike that goes into a shallower Midtown Station. Sure it’s not ideal (awful connectivity at Westlake) but it could save billions! Maybe the savings could even fund a branch into SLU and a bit further.
The monorail will be 73 when this line supposedly opens and we will need To replace or restore things anyway. Because there is already an elevated transit vehicle running there, the visual intrusiveness would be mostly negligible.
Weather a single track is enough or not depends on the frequency and the schedule reliability. The east end of MAX was originally planned for 15 minute base service, doubling to 7.5 minute headways peak. So, the passing sections of track were placed based on that level of service +/- a minute or two. A train would leave Gresham and enter single track for 3 minutes, and then enter double track where, during peak operations, a train would be coming the other way.
If you are running trains more often than this, chances are it would be cheaper to reduce the number of switches and just do double track.
It also means very little schedule flexibility. Once you have the sections of double track installed, those sections determine the train schedule.
So, the length of the line isn’t really what determines when it is time to change to double track. Headway and schedule flexibility and reliability are the determining factors.
This section of Link should be extremely reliable as there shouldn’t be any traffic lights or bridge issues or any of that. Frequency is probably never going to be more than every 10 minutes due to the combination with East Link.
“Running out Delridge is cheaper, but that is significantly different than what was originally proposed”
Truncating the line is fine. It’s substituting another project like Ballard-UW for Ballard-downtown that’s legally doubtful, because voters haven’t approved spending tax money for it.
There’s something wrong with the way Sound Transit is going about its business of sticking tracks on concrete posts. Yes, the new tunnel is going to cost plenty, but $7 billion to stick tracks on posts between Lynnwood and Everett? That seems wildly excessive.
Remember Lynnwood to Everett is the same distance as Lynnwood to Seattle if you follow I-5. The Paine Field detour adds even more track to that.
Very glad to see everybody develop the habit of being constantly on the lookout for alternatives in the face of obstacles. One point to think about: with ongoing developments in tunnel boring and bridge building, in some cases we could do like we did with DSTT: dig and use, say, a tunnel, and let vehicles run surface ’til we an afford to extend grade separation.
Still sorry we couldn’t have put the Route 7 in the DSTT via corner of Dearborn and Rainier. It was only $12 million!
But every idea, also important to research technical considerations, like what the boring machine will have to chew through. Or the elevated pillars rooted in. Fourth dimension is sci-fi. Real-world… need all three. Years ago, was told that a very wide diameter sewer main follows the north shore of the Ship Canal through the U-District. Looks like we got under it.
Really am very curious as to what’ll be in front of a Tunnel Boring Machine on its way to the U-District from Ballard. And so it’ll go. But question to STB staff: have we got any “ins” with the world of machinery as we do with finance, as in this posting? I somehow think we’re going to see some serious large-creature evolution as the ST-‘s roll by.
Mark