Freight’s Last-Mile Problem

An interview with Sandra Rothbard

As a reminder: Changing Lanes will begin reserving most of its content for paying subscribers in May. Pledge to pay now, at this link, and enjoy Charter Membership rates of $8 per month or $80 per year; that’s a 20% discount on regular rates, which will begin on 6 May 2025.

---

The box that appears on your doorstep is a miracle of modern logistics. With a few clicks on a website, you performed a Ritual of Summoning, and after a day’s wait, the thing you wanted materialized from the ether, with no effort required from you.

Or at least it seemed to materialize. Outside of your view, your item winged its way through a supply chain that might cross oceans, through a complex network of warehouses and vehicles. It’s a testament to the efficiency of modern logistics.

But this invisible efficiency masks a growing crisis in our cities. E-commerce sales have grown by approximately 80% since 2017, dramatically increasing the number of delivery vehicles on urban streets. These freight trips make up only 10% of those on U.S. roads but generate 50% of the transportation sector's nitrogen oxide emissions and 57% of fine particulate matter.

The health impacts are profound: truck emissions in NYC's metropolitan region alone contribute to more than 21,000 new childhood asthma cases annually, and cities like Toronto suffer nearly 10,000 years of life lost annually from freight-related exhaust. The economic toll further compounds these issues: city streets suffer gigantic efficiency losses from congestion, decreasing productivity, while curbside conflicts between delivery vehicles, transit, cyclists, and pedestrians create a daily battle for limited urban space.¹

To understand the problem, and potential solutions, better, I spoke with my friend, freight specialist Sandra Rothbard. Her expertise in logistics began in graduate school; while pursuing her master's at NYU with hopes of working in affordable housing, she found her passion through an internship and class project on NYC's solid-waste management plan. After graduation, she joined NYC's Department of Transportation's Office of Freight Mobility, a small team responsible for all road-based goods movement in the city. "This was my dream job," she notes, even though "we got zero support." Her team pitched ideas like micro-hubs, delivery lockers, and cargo bikes that are only being implemented now, 15 years later.

Hurricane Sandy in 2012 transformed her career. She managed the movement of trucks bringing emergency supplies and equipment to the city, and her excellence in the role led eventually to her becoming the first director of the supply chain logistics division at the Office of Emergency Management. After leaving government, Rothbard consulted for Sidewalk Labs, where we met, working on different mobility aspects of the firm’s mooted ‘city of the future’. She moved to Amsterdam in early 2020. Her expectations for what her life in Europe would be like were capsized by the pandemic, but despite that disappointment, she worked on zero-emission freight zones across European cities, including Ravenna and Aarhus. Today, based in Atlanta but working internationally, Rothbard runs Freight Matters, her own consulting firm.

We spoke on 4 February 2025. This transcript of our conversation has been edited for length and clarity.

The Problem with Modern Deliveries

Andrew Miller: Good to see you!

Sandra Rothbard: Nice to see you too!

AMM: Right now, thanks to the explosion of internet purchasing and next-day delivery promises, every company that I'm aware of has got to figure out for themselves how to get stuff through the last mile from their warehouses to consumers. Things seem to work well from a consumer point of view. Is there any problem here that requires our attention, or is the system working well and we should just get out of its way?

SR: Great question. It is not working well.

Delivery truck blocking a bike lane. Photo credit: get directly down on Flickr, CC BY-NC 2.0

Same-day and next-day deliveries sort of destroyed everything. The big problem is we have trucks double parking, parked in bike lanes, parked in bus lanes, because everybody wants their delivery now, and a truck driver doesn't have the time to go search for a loading zone that might not exist or might be full. They're going to get in trouble if that package is not delivered by a certain time. [If there’s a problem], it's the drivers and the carriers and their management, they're the ones that eat the cost.

Does everybody need that pair of sneakers tomorrow? Probably not. There are some items that I imagine people choose same-day or next-day because they legitimately need them. But because that's an option, so sure, I want my stuff. As a result, the only way to do this in a timely manner is truck drivers have to double park or park in a fire hydrant zone or what have you.

That's where, at least for the last mile, e-commerce is a big problem. And e-commerce is not just packages from Amazon. It's grocery, pharmacy, all sorts of things. What we need to do is figure out better curb-management strategies.

AMM: It sounds like there's a fundamental mismatch between the demands we're placing on the system and the infrastructure that's available to support it.

SR: Absolutely. Our urban infrastructure was never designed for this volume of deliveries. Many of the buildings we're still using were built 100 years ago, long before anyone conceived of the e-commerce model. At the same time, we've been redesigning our streets to prioritize transit, cycling, and pedestrians—all good things—but often without considering where the delivery trucks are supposed to go.

A System of Unintended Consequences

AMM: I want to drill down a bit more on the problem statement. If I understood you, the problem is that because the system puts the impetus on the driver to solve the problem—they're the ones that will be punished if they don't hit their targets—they have no incentive to do things in any way other than 'quickly'. What are the problems with this that you aren't seeing if you are a consumer and all you see is that your package showed up?

SR: What you don't see are the safety challenges, the emissions concerns, the congestion… or you see it, but you don't realize the cause. It's hidden in plain sight. You don't recognize that you are part of the problem. You can change your habits, whether it's consolidating your orders or sending it to a delivery locker so a truck can drop off all these items in one trip.

It's better when that trip is replacing a car trip. One of the biggest challenges is there's a thinking of "Well, isn't a truck delivering better for the environment, for safety, whatever, than if I drive to the store?" But the matrix of different what-ifs is so large. It depends on if you're going to the store by transit or walking or cycling. That's very different than going to pick up an item in the store by car. What is that truck replacing?

You're going to get three different boxes for the three different things you ordered in that one order, because this one's coming from this warehouse, this one's coming from this warehouse. People get frustrated but don't realize that's what's happening.

If you're ordering something like a single razor, it's coming in a big box. It's a lot more wasteful and less sustainable than if you go to the store and buy that one razor and put it in a reusable tote. But did you drive to the store? And now it was one car trip with one person in it, versus one truck trip that's also serving multiple addresses.

And now that packaging, if you had delivery at home… now it has to leave at some point. So now you have another truck that needs to come and take all your recycling away, throw all your waste away, because now you have so much more at home. We've generated so much more cardboard waste or bubble wrapper, whatever the item is being shipped in.

And then you also have the reverse logistics. You go to a store, you need a new T-shirt, you're gonna try it on in the store. You're gonna buy the right size. If you're ordering it online, you're gonna buy it in a couple of sizes because you're not really sure what size you are, and the returns are free for most of these stores. So you're gonna try them on, and then you're going to ship it back. So that also creates more truck trips and vehicle miles traveled.

AMM: I've noticed that some delivery companies treat parking tickets as just a business expense.

SR: Exactly. If a truck is double-parked, they incur a fine. It's a cost of doing business for these e-commerce carriers, and at the end of the year, these companies cut a deal with New York City. It's called the stipulated fine program. They say, we know this is going to be an administrative nightmare for you, New York City. So instead of us going through every single thing individually, we will come up with a deal. Let's haggle a little bit, and we'll pay a fraction of that, but we'll just cut you a check. It's just the cost of doing business.

We should consider freight like public transportation. I like to say that all our stuff is in the back of the trucks, right? The clothing we wear, the medicine we take, the food we eat, you know, the random nonsense from Amazon that we buy. And every time we talk with government about, why didn't you consider loading zones, or why aren't you doing this, that, or the other for trucks? The thinking is, well, that's private sector. It's not our responsibility. Actually it is government's responsibility to manage the curb. But we consider freight the private sector.

Meanwhile, government does a whole lot of work and support for people's personal private vehicles. All these EV charging stations and all these other things to try to get folks out of their internal combustion engine vehicles. Well, we don't have enough subsidies to convert trucks to electric. Sure, that's the private sector, but almost all trucking fleets in the US have fewer than five trucks. The majority of those are a single truck. These are mom-and-pop companies. UPS can pay for a transition to electric, but Bob, who owns his own truck, he can't afford it.

Why are we doing this for people's personal vehicles? There's alternatives. You can bike, you can walk, you can take the bus or the train. We do not have alternatives for trucks yet, right? We're getting there with cargo bikes. And there's some other things too. But you know, the government treats this like it's wholly private, and it is, but it really is far more public than someone's personal, private vehicle. We need to treat it that way.

Why Obvious Solutions Fail

SR (continued): I have a real big problem with Complete Streets, because it's incredibly incomplete. So many Complete Streets strategies do not include freight, do not include last-mile delivery and trucks. We talk about having a safe street design that is safe for cyclists and pedestrians and, of course, drivers and transit passengers, but we don't talk about trucks.

AMM: So what would a complete street that takes goods movement into consideration look like? How would that be different?

SR: My pie in the sky, crazy vision is the curb should only be for pick-up-and-drop-off of goods and passengers. Absolutely no parking.

AMM: That is—

SR: Wild and not feasible in the slightest.

AMM: My book [The End of Driving, pre-order here] is coming out later this year. One of the things that we advocate for in the book is cities need to completely reclaim the curb and do pick-up-and-drop-off zones, because absent that, fleets of robotaxis are going to make streets undrivable. One of the things we proposed to do at Sidewalk Labs which I still think was right was, you don't get the curb for free. Maybe it's a small fee, but you can register in advance, you pay it, and the curb is yours for a particular period. It's there for you to pick up and drop off whatever it is you're picking up or dropping off. If you just let people idle at the curb... that way lies madness. It's bad now in some areas, it's going to get even worse. It'll become insupportable.

SR: One of the biggest challenges is, even if you do have a loading zone on a certain street, a truck might still double-park if it gets them closer to where they need to be. They're in a rush. They have to get this delivered now, right? And so it's not worth it for them to park even 50 feet away, because then they have to put their items on a hand truck and then schlep it down the street, even heavier items or fragile items, right?

We created dedicated truck loading/unloading spaces by removing car parking, but when surveying, we found trucks weren't using them. Sometimes personal vehicles or other trucks occupied the spaces due to poor enforcement. But other times, drivers told us that while the space was nice, their delivery point was 50 feet away. Given their need for speed and the challenge of moving heavy items on hand trucks up curbs in various weather conditions, they opted to double-park.

To be clear, delivery drivers don't want to double-park. It's dangerous for them too. They don't want to park next to cars or in bus lanes, bike lanes, or bus stops. It's a very dangerous job, but when they must move quickly and lack better options, they'll take these risks.

Image courtesy of Sandra Rothbard

SR (continued): Technology isn't always the answer, though it can be great for curb management. I worked on an EU-funded project to digitize curb regulations and create a delivery driver app [for cities like Strasbourg and Toulouse]. Everyone was excited about features like reserving loading spaces online, but as the freight expert, I saw major problems.

Reservations don't work except for predictable deliveries like grocery stores. For UPS drivers making quick stops amid unpredictable congestion, they can't know exact arrival times. Variables like whether someone answers their door or elevator access can change schedules completely. Reservations don't suit many delivery types. Also, many delivery companies either prohibit personal phones or provide specific devices, complicating app usage.

We need adequate curb space for deliveries, with careful consideration of location. In Manhattan, I've seen multiple trucks from the same company on one block simultaneously—partly due to lack of consolidation—one truck from the airport, another ground shipment from a distribution center, another express delivery. But they double-park because of insufficient loading zones or delivery windows.

What Victory Looks Like

AMM: We've described the problems with the status quo. If we do all the things that we should do, how will the world look different? What would victory look like for you if New York or Atlanta was doing it the way that you think that they should?

SR: The difference is having more options when you're ordering something online. “Oh, there's more choices. Now, when I check out, I can deliver to a locker, I can do consolidation. I can say, 'Yeah, you know what? Just bring it to me when the rest of my neighborhood or my street get something delivered.'” For drivers and carriers, it's “Oh, there's dedicated space for us to load and unload at the curb. Somebody in government has thought about us and their street design has considered our needs”.

AMM: Let's talk about micro-hubs. They make intuitive sense. Why not have a warehouse where all goods for a certain radius around it come in, then the goods fan out from there? Maybe on foot, cargo bike, or even delivery robot. [Microhubs scale in size from 1,000 to 10,000 square feet.] But I wonder: what's the incentive? If I'm Purolator or UPS or DHL, why should I put that step in my process outside of my control?

SR: The incentive is, if you're Purolator and have a truck full of packages with many stops, going to a micro-hub means one stop. They're unloading that whole truck, and everyone else handles the last mile. (Lockers are a type of micro-hub.) We were also looking at urban consolidation centers, which were much bigger and had other uses inside.

The urban consolidation center model can be difficult because it adds an extra touch point, but a micro-hub essentially replaces multiple consumers with one consumer and one stop… much faster for carriers. They don't have to worry as much about security and theft. The package will be protected in a locker or behind a door, or someone will be there physically to collect it. Otherwise, you risk porch pirates stealing packages left on front porches.

That raises another problem… when folks don't get their deliveries on time, the right party doesn't necessarily get the blame. Who suffers the consequences? That's another advantage to dropping everything off in one place. It's less stressful if micro-hubs are designed properly in locations with appropriate loading/unloading space, versus a driver having to find parking at up to 10 different locations.

AMM: How do you get firms to start thinking about alternatives? It's much easier for hipsters in Brooklyn with the gin they’ve distilled to just send a truck into Manhattan to deliver it. How do you persuade them to use something more complicated, like cargo bikes and ferries?

SR: It's funny that you said gin. I usually say artisanal mustard when I'm talking about some hipster nonsense.

Cargo bikes can go quite fast. I have an e-bike, that's mostly how I get around Atlanta. I can go up to 25 miles per hour, usually hovering around 20. In New York City, they just limited cargo bikes to 15 miles per hour. [And they] can go pretty far on a single battery charge. But typically distance doesn't matter; what matters is the density of where your customers are. You don't really have to go more than a mile for many of these to make sense. But you have to see the bottom line [to decide].

Cargo bikes [are] not always the right option, but [they’re] really great in places where trucks just physically can't get to, like pedestrianized areas, or where it's just not worth it for a truck to go. [But they only] make sense in a bike lane only if they're a certain size, speed, and if the bike lane is designed appropriately. With a two-way bike lane, if the bike doesn't have a tall design or box in the back, that works fine. But some bike lanes have bollards in the middle to prevent cars from entering, which some cargo bikes can't navigate around.

So there's many reasons why one should choose a cargo bike versus not, but we need a shared system. Just like retrofitting a truck or getting a new electric truck, cargo bikes are expensive.

AMM: Let's talk about building design. We've got buildings we're still using from 100 years ago that clearly weren't designed for modern delivery needs. What should we be designing for that we wouldn't have considered 50 or even 10 years ago?

SR: Newer buildings have not only better package rooms but also refrigerated storage space. If you order groceries and aren't home, they need climate control. That's been around for a while in luxury buildings. There are technologies we saw at Sidewalk Labs where you go into a room with simple shelves and special computer vision cameras that track where packages are placed. You don't need specific-sized lockers… put packages wherever they fit on the shelf, scan the barcode, and the computer knows it's on the second shelf, third box in. When residents come home, it tells them exactly where their package is, maybe with a light indicator.

For all our designs—I hate the term “future-proof” but I'll use it here—something's going to change in five years, so what if we have to retrofit or redo it? You need something flexible enough that if requirements change, we can adapt these spaces.

What Cities Can Do Now

AMM: Based on your experience in New York, Amsterdam, Atlanta, and other cities, who's doing it right? What city should everyone pay attention to?

SR: That's such a hard question.

AMM: Well, let's put it another way. If a mayor asks what one thing they could do to meaningfully improve goods movement in their city, what would you tell them?

SR: Make sure your Department of Transportation or Public Works has someone—or better yet, multiple people—dedicated solely to last-mile delivery or freight generally. That's the single best thing you can do. Currently, you have to hope other folks working on transportation or business plans include freight, which they typically don't.

We need to educate staff and ensure dedicated freight expertise. Some departments have significant bike or bus groups, but freight encompasses all these areas. Covid helped people recognize that while we can manage without moving people, we're in trouble if goods don't move. As a freight planner, I need to understand bike lanes and cargo bikes, transit operations, bus lanes, cargo hitching [i.e., moving goods via public transit, under buses or on ferries], land use, warehousing, zoning, and multimodal deliveries beyond roads. I need to understand rail and maritime movement.

I don't need to be an absolute expert, but I need broader knowledge than typical transportation planners who focus on one area. People often say "Oh, you work in freight. How niche!" But my world is broader than theirs. I have to know all these aspects. If you ask the average bike planner about cargo bikes and freight, many will say they don't consider that.

AMM: If I've understood correctly, this happens because the private sector excels at solving certain aspects, so we let them handle it, missing other facets because the visible parts seem managed by the market?

SR: Yes, which is why having a freight office bridges that gap between industry and other transportation aspects, and between industry and government generally. You need people solely focused on this broad—not niche—area. You need people creating freight plans and ensuring freight has appropriate consideration in broader comprehensive plans, zoning plans, and transportation plans. It's often missing.

It's about educating people too, not just picking someone random from the bus group because you desperately need to write a plan. We need the right people, more freight education in graduate planning and engineering programs, so people enter these positions with expertise. Creating a freight office within transportation, planning, or public works departments is the starting point. Without that dedicated coordination within and outside government, you won't succeed.

---

I’m grateful to Sandra for speaking with me about the last-mile delivery problem. As our conversation shows, it’s an inconvenient truth that one-day delivery works for consumers but doesn't work well for cities. The convenience of fast delivery comes with considerable costs that remain largely invisible to those who benefit from it.

Consistent with other positions I’ve taken at Changing Lanes, I think that in this arena, technology like cargo-bike sharing or camera-vision package storage can help, but it’s downstream of other solutions: introducing new policy regarding curb use, redesigning streets to facilitate loading docks, and establishing dedicated freight offices. It will take work to solve the last mile problem… but the first step will be recognizing that it is a problem, and unless cities act, it won’t be solved.

Respect to

Jennifer Morales

and

Jannik Reigl

for feedback on earlier drafts.

1. I’ve drawn the statistics in this and the preceding paragraph from this report by the Environmental Defense Fund and this article by Fried et al. (2024).