Any regular transit rider coming home from Bellevue, in a bus, in an HOV lane on I-405 southbound, knows well the feeling of moving 0-5 mph. Granted, it’s not always like this. There are certainly some days where it zips by traffic at nearly 60, while some other days it takes over 20 minutes just to get to I-90.

While unreliable HOV travel times are already quite frustrating, very often traffic flow in the HOV lane ends up being as bad or worse than in the general purpose lanes! Why is this tolerated by WSDOT, when the whole point of an HOV lane is to flow faster as an incentive for people to carpool or take transit? Why would anyone want to do this if they are just going to get stuck in the same traffic as if they drove alone?

I-405 south of Bellevue is getting the same type of express-toll lanes as to the north, but not until 2024. You would think that making the current HOV lanes HOV3+ would be a natural precursor to the eventual ETL extension, but WSDOT would not agree.

But how many service hours might be saved if HOV3+ were in place on I-405, in particular from Renton to Bellevue? In this calculation, I’ll assume that traffic in the HOV3+ lanes always moves at 45 miles per hour or better. That is the standard which WSDOT attempts to maintain for the express-toll lanes generally, and is also the point at which WSDOT says it will consider upgrading HOV2+ lanes to HOV3+ (though clearly that doesn’t seem to mean anything in practice).

This calculation is going to be extremely rough for many reasons. Not only are actual vehicle speeds variable, but it’s also hard to figure out how much of the time buses use the HOV lane, since (on I-405 in particular), it often varies not just by route, but by driver. 560 operators often pop into the HOV lane between Kennydale and Newport Hills freeway stations (possible because each direction is only 3 lanes wide), while on other routes (select 566 trips and Metro routes) operators tend to stay in the right lane, not wanting to risk missing the next exit.

My analysis assumes trips serving these freeway stations do not utilize the HOV lane until north of Newport Hills freeway station. It also only calculates based on two major congestion cases (I-405 south from NE 6th to Newport Hills in the afternoon, and I-405 north from Southport Drive to NE 6th in the morning), to approximate time savings for the most consistent slowdowns (southbound from NE 6th to I-90, and northbound from SR 167 to Coal Creek Parkway SE). Outside these scenarios, the HOV lane tends to be effective. I am most likely to underestimate the savings, since I only count cases where the time savings is reasonably clear. As such, certain routes are excluded from the analysis. Route 342 only operates in the reverse-peak direction. Routes 111 and 114 make relatively closely-spaced right-hand exits on the freeway. And in the northbound direction, routes 560, 167, and 952 are also excluded since the northbound analysis considers only congestion south of Coal Creek Pkwy. Though the latter will list savings for these routes as “0,” there is nearly certainly some non-quantifiable time savings on these trips from HOV3.

Below are the approximate minutes saved per route, per direction, per day. These come from scheduled stops on ST 560 SB and ST 566 NB, and the number of minutes saved if traffic on I-405 moved at 45 miles per hour. ST 560 from Bellevue Transit Center to Newport Hills Freeway Station would take no longer than 9 minutes, and ST 566 from Southport Drive to Bellevue TC would take no longer than 18 minutes:

560566567167952ST TotalMetro TotalTotal
Total3992702928201 mins57 mins258

*Extrapolated from nearby 560/566 trips **Not able to quantify HOV3+ time savings

Converting minutes per weekday to hours per year (assuming approximately 255 business days per year), then Sound Transit saves 854 annual service hours, and King County Metro saves 230 annual service hours. While only adding up to about $150,000 in annual savings, this is likely to be a signficant underestimate.

39 Replies to “The cost of HOV-2 on I-405”

  1. Yeah, doing this would be huge. My guess is if they changed HOV2 to HOV3 on I-5, it would make an even bigger difference. The math gets even more complicated there. Peak hour peak direction buses run in the express lanes, which allow them to go faster. But those only cover part of the ride for some buses. Buses like the 41, which run all day, suffer with horrible traffic in the evening, opposite the express lanes. Try and take the bus downtown in the evening and it takes forever. A bus like the 512 is even worse. In the evening, it gets stuck going both directions — from the end of the express lanes to Everett heading north, and from about 145th to downtown heading south. The 510 has a similar problem, even if part of that is dead-heading (it still costs money).

    That is all for the north end. The south end can be worse, as there are no express lanes. Changing HOV2 to HOV3 on the roads that need it would be the cheapest, easiest way to significantly improve transit in the state.

    1. With I-5 north of downtown Seattle not having any HOV lanes at all when there are express lanes, that seems like an opportunity. There was a post a while back about the serious technical challenges of putting HOV lanes on the mainline, but making half of the express lanes HOV2 would certainly help. Making the whole thing HOV2 would help even more, though that will really make lots of people mad. If that were done though, HOV2 might end up being enough to keep vehicles moving at 45 or better through the express lanes, since these are 4 (I think?) lanes rather than 1.

      1. With Link at Northgate (very soon), I wouldn’t bother worrying about the lack of HOV lanes on the mainline between Northgate and downtown. There simply won’t be any buses doing that. The Community Transit express buses to downtown will only go peak direction (thus using the express lanes). My understanding is that they don’t even deadhead, but rather, just layover downtown. The all day Metro and Sound Transit buses will be terminated at Northgate. No sense fighting for something that won’t even be an issue in a few years.

        You could probably say the same thing about HOV-2 to HOV-3 between downtown and Lynnwood. Once Lynnwood Link gets here, it won’t matter. But that is further away, and it makes sense to do the whole road. You could, theoretically, have HOV-3 the whole way now, but then transition to a mixed set once Lynnwood Link gets here (HOV-2 up to Lynnwood, then changing to HOV-3). That seems silly though. We are still likely to have vanpools and bigger carpools — my guess is the case for HOV-3 will remain, even after the buses don’t use that section.

        North of there, it makes sense to have HOV-3, even after Everett Link is complete. Express buses to Lynnwood (serving places like South Everett Park and Ride) should be able to go fast, saving riders a lot of time.

      2. «technical challenges of putting HOV lanes on the mainline» How technically challenging is it to paint a solid white line and some diamonds?

  2. Also worth noting – when you’re driving your own car, you only pay for bad traffic on days when their actually is bad traffic. But, when you’re operating a bus, you effectively have to pay for bad traffic every single day, even on days when traffic is relatively light. This is because the schedules and driver/bus assignments are written in advance. If the traffic is good, the bus may finish early, but that just means a longer layover. Either way, the bus is still tied up, and driver is still getting paid for the full scheduled duration of the run.

    It gets even worse if you start thinking about the passenger experience for a route like the 566. Either people getting on in Renton will have very unpredictable wait times (depending on the whims of traffic along 167 northbound or 405 southbound), or the bus has to have a long mid-route layover at Renton Transit Center, leading to unbearably long travel times for thru-riders when traffic is light. Either way, it’s a lose-lose situation, and it’s impossible to help one rider group without hurting the other.

    1. This is exactly right. 566 southbound schedules a lot of time to get from Overlake TC to Bellevue, just because traffic on 520/405 is so unpredictable. Often the bus is more than 10 minutes early and just waits at BTC, and on bad days (not too often because of how much time is planned) it’s 10 minutes late.

      167 northbound traffic has indeed been killing 566 reliability in the morning. I just took a 566 that was 18 minutes late when it left Renton. It seems like Renton to Bellevue should just be its own route, like Kent to Bellevue is (and Renton to Seattle is). Auburn to Bellevue should just be variations of the 567, not 566.

      What’s so frustrating about the 566 is that traffic on 405 is bad enough for buses. But what makes it even worse is the way ST chooses to route the buses, making so many people sit through the city of Renton tour, and making people starting in Renton have to wait for the bus to get through 167 traffic first.

      1. The problem is – in order to make everyone happy, ST could have to run three separate route. One from Bellevue to Renton, one from Renton to Kent, and one from Bellevue to Kent, skipping Renton. That means spreading service very thin, which means low frequency, half-empty buses, and either extremely high subsidies per rider or extremely high fares.

        The red carpet every-trip-pair-gets-a-separate-bus model is pretty much what you among corporate commuter shuttles, such as the Microsoft Connector. Unfortunately, it is not cost-effective for a public agency like Sound Transit to run route like that. (Microsoft is able to justify a high subsidy per rider that Sound Transit would not accept, because it helps with employee recruitment and retention, plus the extra productivity from people who choose to get work done one the bus).

        For a public agency, in order to be cost-effective, you have to combine multiple trips into a single route. In order to do that without driving passengers away, you need reliable service, plus the physical infrastructure to make quick stops to load and unload passengers without getting off the freeway and waiting at a bunch of extra stoplights and traffic for every single stop. These are the issues that I-405 BRT is meant to address.

        In the case of the Overlake extension and the backups on the 520-405 interchange, one can only hope that the 566 will be truncated in Bellevue, once East Link opens, allowing southbound trips to at least leave Bellevue reliably on-time. The latest ST operations plan I saw shows the 566 continuing to Overlake indefinitely, which seems absolutely crazy. During the limited hours that the 566 is actually running, I don’t see how the 566 can possibly get from Overlake Transit Center to Bellevue Transit Center faster than Link under any kind of remotely realistic traffic conditions. Just the stoplights getting on and off the freeway is already enough time for a Link train to be most of the way there.

      2. These are all meaningful tradeoffs to consider. I have some ideas (I’ll probably do a page 2 in the coming weeks about them, and a basic service hour analysis to figure out what’s possible). But the way it is now just seems like the remnants of last decade’s very coverage-oriented system of routes that made 4 or more sets of off-freeway stops in every major city on the way to either Federal Way or Puyallup, just incrementally and in a very ad-hoc way made to be more commuter-oriented (most visibly by adding the 567, but not really restructuring the rest of the routes in a comprehensive way). Because of this, there are weird service patterns with Auburn getting the slow bus, and the 566 still trying to be all things to all people. It would be like if the 592 stopped in Tacoma and Federal Way on the easy way to Seattle. It wouldn’t really work well for most people, even if it helps some people get to Federal Way.

      3. The problem is – in order to make everyone happy, ST could have to run three separate route. One from Bellevue to Renton, one from Renton to Kent, and one from Bellevue to Kent, skipping Renton.

        I think Alex is right, though. The route to skip is Renton to Kent. Metro already provides that, even if it is slower. There just aren’t that many riders heading between Renton and Kent. Of the 670 riders heading north every day, about 75 get off at Renton. That is not that many. The vast majority of people are headed to Bellevue.

        As Alex wrote, I would tack on the Auburn piece to the 567. Along with the loss of service between Auburn/Kent and Renton, it would mean that Auburn/Kent riders would have a hard time getting to Bellevue in the middle of the day. So be it. The 566 gets somewhere between 10 to 20 riders in the middle of the day (and that includes everyone). If it was run by Metro, it would immediately become a peak only bus (since it performs so poorly the rest of the day). There just aren’t enough midday riders from Kent and Auburn on the 566 to worry about.

        From a service perspective, I would start the discussion from the current levels. The 566 runs 28 times a day (one direction). Shifting the Kent/Auburn section to the 567 saves a lot of money, because the 567 runs only ten times a day. You also save a little bit since the 567 avoids going to Renton (which means it stays in the HOV lanes from 167 to 405). With the savings, I would extend the hours of the 567 (maybe add 3 or 4 new runs each direction). As mentioned, I wouldn’t run the 567 at noon (those riders have to use Metro at least part of the way). I would keep the 566 runs about the same.

        There might be some additional shifting. The 566 runs a lot during rush hour because of crowding. But some of those folks came from Auburn, and some came from Kent (if they missed the 567). You may not need as many 566 buses at the peak of peak, which means that you can shift some service from the 566 to the 567. Riders on the 566 won’t care (frequency is already really high) while 567 riders would see a big increase in frequency (over the current 20 minutes). It might make sense to have two versions of the 566 — one from Kent (the existing 567) and one that has the Auburn tail. That way, during rush hour, you could get ten minute headways from Kent, without spending a fortune. My guess is if you did all of that, the vast majority of riders would come out ahead, for no additional money. This, in turn, would increase ridership.

        Killing off the Overlake section saves money, but it isn’t clear what picks up the gap. There aren’t a ton of riders using that Overlake stop (about 10% of the riders). You could simply ask them to switch to the B. The B is much slower, but goes to more places — it might be OK for most of the riders. Instead of fast trip to Overlake followed by a long walk (or a transfer) they transfer to the B at Bellevue TC, and get right to their stop.

        I think there is merit in an Overlake TC to Metro TC “express” (since the B is pretty slow getting there) but I’m not sure what it would look like. It could be a bidirectional rush hour line (since there are commuters heading to both Bellevue and Redmond) but then you struggle the same way with reliability. With the B as backup, I’m not sure how big of a problem that would be though. In the morning it would be no big deal. You catch the bus to Bellevue TC, then you look to see if the express is there (or about to be there). If not, you take the B.

        It would be in the evening where things are risky. You do the same thing, and take the first bus, but find that you get there just as the 566 or 567 is leaving. If there is enough frequency on both it isn’t the end of the world (especially compared to a long layover at Bellevue TC) but you would need good frequency on both. It seems possible — the 566 is frequent, and the 567 could be as well — but it wouldn’t be a slam dunk.

        Another alternative is to just live with the B and truncate both routes. The combination of no ST service between Kent/Auburn and Renton as well as no one-seat ride to Overlake would hurt about 20% of the riders. But they could still make the trip, while the other 80% would have a much better ride — with better frequency and a lot better speed.

      4. Of course, eventually all these problems go away. The train goes to Overlake and 405 BRT replaces Renton to Bellevue service. At that point, I would definitely focus on the 567 (with an Auburn tail). I would not go to Renton, but make only freeway stops between Kent and downtown Bellevue (where the bus would terminate). Those headed to Renton from Auburn/Kent would either backtrack from 44th, or rely on Metro. Metro (or even ST) might add an express version of the 153, but my guess is wouldn’t have enough riders to justify it. By then 405 will have HOT lanes the whole way, which would make getting from Kent to Bellevue pretty fast, while detouring to (or transferring at) Renton would be a significant delay for the vast majority of riders.

      5. RossB,

        I am thinking very much along those lines. Midday 566 service is currently hourly only to Renton TC (not Bellevue) and has a timed *and waiting* connection from the 560 (so the 566 will wait for the 560 if it’s late). I think this is smart, but riders might think it’s harder to get to Bellevue than it really is. Those ridership numbers do seem really low though.

        For the 567, making a variant that serves Auburn is a must (call it the 568), but haven taken the 567 from time to time, I find that the bus fairly often gets completely full with passengers from the Sounder train (even the huge MCI coaches). So trips where this happens are the perfect candidates to keep the trips as 567, and it makes the most sense to put 568 trips in between these because then it makes frequent 10 minute service from Kent, faster and more frequent than today service to Auburn (20 minutes), and provides some spillover capacity for Sounder riders who didn’t make the 567 because it’s full (they can wait 10 minutes for the 568).

        Sure, this will be moot when Stride opens, but that’s not for 5 years (or longer depending on how 976 pans out). That’s a lot of time for people to learn how painful it is to take the bus here.

      6. I think one option would be to take the same approach as Snohomish County does with the 510, 511 and 512. This is a clever routing, in my opinion. The 512 – the bus that goes from Everett to Seattle and stops in Ash Way, Lynnwood, etc. along the way — just doesn’t run peak direction during rush hour. You have an express bus from Seattle to Everett (which skips Lynnwood). You have an express bus from Seattle to Lynnwood. But at 5:00 PM, if you want to get from Lynnwood to Everett, you just have to catch a Community Transit bus. This makes the express (from Everett) a lot faster when it matters most.

        I could easily see the same approach here. Run buses from Auburn and Kent to Bellevue rush hour direction. Also run buses from Renton to Bellevue. As mentioned, that actually would result in better balance. The Renton to Bellevue bus wouldn’t be so crowded, while the Auburn/Kent bus would have a lot more riders and thus better frequency.

        Then, when rush hour is over, you run the existing 566. I didn’t realize it only ran to Renton. That makes the change even more affordable and easy. Keep the same midday routing, but modify the rush hour buses. You probably end up with additional frequency where it is needed most (in Kent and Auburn) when it is needed most (rush hour).

        The 510 skips Ash Way and Lynnwood because it is very time consuming to serve Ash Way. But serving Renton is far worse. It crazy to ignore the brand new ramps. It is a huge delay for Kent/Auburn riders, and even slows down Renton riders since they want some semblance of a real schedule. This would be a big improvement for the vast majority of riders.

        I agree, we should change things now, and not wait five years.

      7. I would keep the 566 to Overlake for three more years until Link opens, and truncate it then. The B line is just too slow to function as a valid replacement. Link will be much faster, and much more reliable, and is a completely different situation.

      8. I would keep the 566 to Overlake for three more years until Link opens, and truncate it then. The B line is just too slow to function as a valid replacement.

        Yeah, but that favors the few to the detriment of the many. About 10% of the riders are headed up to Overlake. The Bellevue riders are hurt because of the service cost to go to Overlake, and the delay in having to wait for the bus to get there (when traveling southbound). The same is true for serving Renton on the way to Bellevue. About 10% of the riders are going from Kent or Auburn to Renton, and they too have an alternative.

        Thus about 80% of the riders are getting significantly slower and less frequent service while the other 20% avoid a slight delay (and/or transfer).

        I’m not saying I would make any changes either. It is far easier, politically, just to change the subject, and wait for East Link and I-405 BRT, both of which arrive at roughly the same time.

  3. Are we talking about HOV3 lanes or HOT lanes? The latter are inherently classist and exist along part of this corridor, ergo the confusion. SOV drivers simply paying a fee to use HOV lanes is not something we should encourage.

    1. He is talking about having HOV-3 lanes before we have HOT lanes. To quote the article:

      I-405 south of Bellevue is getting the same type of express-toll lanes as to the north, but not until 2024. You would think that making the current HOV lanes HOV3+ would be a natural precursor to the eventual ETL extension, but WSDOT would not agree.

      The HOT lanes are coming, like it or not. Alex is simply suggesting we have HOV-3 lanes before then. Given the cost of actually converting HOV-2 to HOV-3, this is a very good idea.

      In general, this is just a variation on what we need to do everywhere. If an HOV-2 lane experiences major congestion — if it can’t live up to its original promise* — then it should be bumped up to HOV-3. This really should be a political priority, but it seems to be ignored every year.

      * I forget the details, but there is something about the HOV lanes needing to flow at 45 MPH before the feds will fund them. It is clear that they aren’t operating properly in many areas (I-5, I-405, etc.).

    2. Classist, schmassist. If a toll on SOV users keeps the buses running at optimum speed, let Donald Trump ride The Beast down it every morning. Life isn’t fair from whenever you mark its beginning.

      1. But HOT lanes run the the risk of preventing the busses from running at optimum speed. Raising the toll will not deter some users from using the lanes, thereby effectively causing a semipermanent level of SOV congestion on the HOV3 lanes.

      2. You’re obviously not a driver of I-5. Half the cars in the HOV lane have one person in them. There is NO enforcement. Calling the “Hero” line a futile gesture.

        At least the toll lanes can be monitored by license plate.

      3. For clarity. Once the State has a direct interest (e.g. “toll revenues”) in keeping the SOV’s out of the HOT lanes, they will enforce them.

      4. At least the toll lanes can be monitored by license plate.

        Can they though? What if I get a Flex Pass ( and then set it to “HOV mode”. It seems to me that the cops are in the same spot, only worse. They have to somehow monitor that I’m not paying the toll (which seems possible, but at best it is an extra step) AND then determine whether there are three people in the car. With a regular carpool lane, a cop just looks at each and every car to see if they have enough people.

      5. I don’t think the flex pass itself is a major issue. When set to HOV they show a red square on the front of the transponder. If a cop is looking at the front of the car, they’ll see pretty simply if it’s set to HOV or not.

        In practice, though, I don’t know how much enforcement happens on the actual road just because it is hard. I see a lot of enforcement on exits – the HOT exit at 6th street in Bellevue very often has enforcement right as people get off 405. I’ve also seen a trooper standing on entrance ramps where there’s an HOV lane and just flagging people down as they pass by.

        As for monitoring by license plate, I’ve been told by Good to Go that they have no records when a pass is set to HOV. At least, they can’t tell me when my pass was set to HOV and passed through a 405 toll point. Perhaps it’s stored somewhere, but it wouldn’t surprise me if it’s not.

      6. lack of enforcement is the problem.
        on both i90 and 520.

        onramps get enforcement perhaps once per year.
        lanes get enforcement, never.

    3. Most HOV lanes in the region are HOV-2. WSDOT put HOV-3 only in a few extreme cases like the eastern approach to the 520 bridge where the right of way was very narrow and they eked out an HOV lane from the shoulder lane. it may be gone now with the 520 redevelopment. We have long argued for WSDOT to put it on I-5 and 405 but it won’t because it thinks conditions aren’t bad enough to inconvenience 2-person cars. (“Who is the freeway for, anyway? Cars.”) This article is another plea to implement HOV-3, to guarantee that buses can move at at least 45 mph.

      HOT is an overlapping but distinct issue. An HOT lane is essentially an HOV lane plus toll-paying SOVs. Theoretically it shouldn’t matter as long as the number of SOVs is low enough to not slow down the buses. The only mechanism WSDOT has to manage that is the toll rate. The toll was capped at $10 thinking it would never rise that high, but like a hundred-year flood it does happen more than expected and people are surprisingly willing to pay $10 and crowd into the HOT lane. I have no hard-and-fast opionion on HOT lanes but the 45 mph minimum should be guaranteed somehow, and HOT-3 would help.

      1. I think WSDOT is viewing them as tolled fast lanes, with the *extra* benefit that carpools of 3 or more can use them for free. That’s why there are two of them, and (critically) they are tolled based on traffic volume. I do think the toll ceiling should probably be raised to something like $25 though. Having drivers sitting traffic envious of the $20 fast lane, and then seeing Stride buses zoom down every 10 minutes is good marketing for the buses, because it provides an equivalent value (If I take the bus, I get the benefit of a $20 lane for just $3.25!). It also does give a bone to HOV2 carpoolers in that they can effectively pay half the toll since they can split the toll between the two of them (although most probably don’t view it that way), and it gives them a real incentive to bring one more friend to make it free.

      2. Yeah, I agree. I understand why folks hate the HOT lanes (why should rich people get the added public benefit?) but the more I’ve heard about the arguments, the more I figure it is about the best we can do. We live in a regressive tax state — this is likely more progressive than most taxes and fees. Not all of the people who use the HOT lane are rich, either. There are special circumstances where the lane is worth it, even for middle class people (late for a job interview, late to pick up your kids from daycare, etc.). You also have people who can’t possibly take the bus, like truck drivers on a tight timeline.

        Complaining about this because it benefits the rich seems like complaining about a symptom, not the cause. Disparity in income is a real problem in this country, and it causes many problems. For most people, paying a few extra bucks is a pain — ten dollars to just go a mile or two seems crazy. But obviously, to a few (or perhaps, more than a few) it is no big deal. They are willing to pay this every time. But that doesn’t mean that the basic principle (a high cost road for those who really need it, combined with a high occupancy road) is a bad concept.

      3. This seems an awful lot like a “the ends justify the means” argument. I’d argue the ends are the means, and through that lens HOT lanes are incredibly problematic.

      4. This seems an awful lot like a “the ends justify the means” argument.

        Not entirely. The point I’m making is that HOT lanes have merit in a more middle class society. There are various people who are willing to pay more, not because they can afford to, but because it is more important to them *at that moment*. The ideal system would be one in which you pay relative to your means (similar to how some countries ticket you based on your income). But that is true for all sorts of things. To a rich person, a speeding ticket is nothing. Nor are other fees and taxes. HOT lanes have merit, even if extremely wealthy people take advantage of them. The problem is that we have extremely wealthy people — and a system that encourages extreme wealth — not that various public services that are meant to be regulated with fees are essentially free to them.

        I am curious as to how many people use the HOT lanes, and what the income distribution of those people are. That data might not be available. But just knowing if there are some people who use it a lot would be valuable. If so, then you could institute a system where prices go up the more you use it. In other words, if you spend a $100 a month in the HOT lanes, you pay the regular rate, but the next hundred gets charged double, etc. That could be a way to limit excess use, while likely being somewhat progressive (those that don’t mind paying over a $100 a month are probably wealthier than average).

      5. HOT lanes are also a way to raise revenue when the other options are very limited and vulnerable to being taken away.

      6. I’d have to see hard data showing that these one off spur of the moment uses represent a significant portion of HOT users before I believed it.

        Raising revenue by creating preferential use cases is not a good idea. A fair system in that regard would be HOV2 usage in a HOT lane. The moment you raise that to an HOV3 lane you’re punishing lower income brackets in order to squeeze a bit of money out of higher income brackets. That’s classist, and classism should have no place in transportation infrastructure, period.

      7. You’re rewarding lower-income brackets by making the buses fast so they can be a more viable alternative.

      8. you’re punishing lower income brackets in order to squeeze a bit of money out of higher income brackets

        Again, not entirely. Even in a society where everyone had the exact same income and the exact same amount of wealth — a GINI coefficient of zero — you would still have people *occasionally* paying to drive the lane, just as you have people *occasionally* going out to dinner, or buying expensive wine. The occasions when you are willing to pay 10 bucks to get there quicker may be just as special, even if they aren’t “special occasions”.

      9. Again, I’ll believe this occasional use case is significant when I see data showing its significance. To me, it doesn’t pass the smell test.

      10. Come on, man, its ten bucks (at the most). Are you saying middle class people aren’t willing to spend 10 bucks just to get somewhere faster? That implies that everyone who rides Lyft, Uber, and every other taxi is rich. Of course the users skew towards upper income folks, but I know plenty of people who use Lyft and Uber who are middle, if not lower income. They use it because the alternative really sucks, and in some cases, costs them more. If your asshole boss tells you “next time you’re late, your fired”, then you drive in that lane. If you got stuck at work and need to pick up your kid at daycare, you drive in that lane. There are a million and one reasons why someone would pay to drive in the HOT lane. I don’t think we can assume that everyone driving it is rich.

        But yeah, I would certainly like to see the data.

    4. There are also people arguing autonomous cars should have free access to the HOV/T lanes, or electric cars, or ridershare cars. So there are forces trying to dilute it even further.

  4. I’m reminded of the El Monte HOV 3 to 2 trial done in 1999. It was documented by TTI and published by the FHWA.

    The report is here:

    Curiously, key speed figures are missing from the Final Report, as loaded on my phone (Anti-transit sabotage??). The text on Page 13 reports that the speeds dropped from 65 to 20 mph in the mornings westbound and 65 to 37 in the afternoons eastbound. The speed decline and unreliability really created problems for 18,000 daily Foothill Transit riders.

    The issue is not as simple as just peak hours for 405 though. This stretch of road is congested over much of the weekday, and there aren’t nearly that many riders in the HOV lanes. It’s also hard to vary the minimum occupancy during the day with signage once it goes to HOT.

Comments are closed.