A few days ago I happened to stop by the local supermarket during the post-work rush. When I was ready to check out all the regular lanes had long lines. Ordinarily, I wouldn’t mind waiting a few minutes, but on this particular evening I had dinner plans I couldn’t be late for. So I shelled out the extra $6 for the express lane so I could skip the lines.
Pe
rceptive readers will surmise that I made up the previous paragraph. In reality, grocery stores don’t impose congestion charges on their customers. Nor do many other types of private businesses. I don’t think I’ve ever had a customer support line offer to expedite my call for a fee. I’ve heard you can get a table faster at some restaurants by bribing the Maitre d’, but I’ve never seen a restaurant formalize the practice. Generally speaking, when businesses experience temporary spikes in demand, they serve customers on a first-come-first-serve basis; they don’t auction off the temporarily scarce capacity to the highest bidder.
These examples came to mind as I was reading the comments on my recent article about congestion pricing of the highways. One of the striking things about the congestion pricing debate is the stark divide between elites and the general public. Prominent intellectuals from across the political spectrum—from free-market transportation experts to left-wing bloggers are supportive of the idea. In contrast, the readership of Ars Technica, like voters generally, is overwhelmingly opposed. And their criticisms were not limited to the privacy issues that were the focus of my article.
Many supporters of congestion pricing chalk this up to voter ignorance. They assume that people will like congestion pricing once they have a chance to try it. But the supermarket example should make us skeptical of that conclusion. The grocery business is an intensely competitive one. If it were true that people could be won over to this kind of scheme once they had a chance to try it, you’d expect some entrepreneurial grocery store owner to give it a try. Yet I’ve lived in half a dozen different metropolitan areas and I’ve never seen a supermarket that utilized congestion pricing on its checkout lanes.
I think there are two reasons that people hate congestion pricing. First, we have strong and sophisticated social norms, cultivated since we were young children, for waiting in lines. This bit of self-organization is extremely important for the smooth functioning of civil society. We see waiting your turn as an obligation we have to one another, and therefore not as an obligation that a supermarket or transportation agency can waive in exchange for a cash payment. I suspect customers would see people using a tolled checkout lane as breaking an implicit social contract.
More importantly, customers would be suspicious that the supermarket was deliberately under-staffing the free lanes to gin up demand for the express ones. And this wouldn’t be a crazy suspicion! In the low-margin grocery business, it would be a pretty effective way for a manager to pump up his short-term profits, while the long-term harm to the store’s reputation would be hard for the corporate office to quantify.
This latter concern seems particularly relevant to the case of toll roads. The revenue-maximizing pricing schedule is not the same as the congestion-minimizing schedule. An effective congestion-pricing scheme might generate relatively little revenue if people shift their driving to off-peak times (which is the whole point). The operator of a monopolistic toll road will face a constant temptation to boost revenues by limiting throughput on free lanes and jacking up the off-peak toll rates. The widespread voter perception that they’ve “already paid for” many tolled roads through other taxes isn’t exactly right as a matter of fiscal policy, but I think it’s based on a sound intuition: there’s no reason to think the political process will set tolls in a way that’s either fair or economically efficient.
I think there’s a lesson here about the relationship between free markets and individual liberty. Free markets promote freedom by enhancing choice and competition. Consumers patronize those businesses that do the best job of satisfying their preferences, and businesses that fail to satisfy their customers eventually go out of business. Yet here’s a case where, in the name of free markets, advocates of congestion tolling are advocating the use of a market mechanism that private firms in actual competitive markets rarely use.
I’ve written before about cases where people become so enthusiastic about the “market” part of free markets that they pay short shrift to the “free” part. Markets are a means to the end of satisfying consumer preferences. Markets cater to consumers’ actual preferences, not to the preferences economists think they ought to have. And government services should follow the same philosophy. The fact that some economists have decided that a particular consumer preference is irrational isn’t a good reason to disregard it. And doing so certainly has nothing to do with free markets.
To be clear, I think there are limited circumstances where congestion tolling makes sense. There’s a lot to be said for the emerging HOT lane model, which has the virtue that the adjacent free lane checks the ability of the tolling authority to charge monopolistic rates. With budgets tight, tolling may be the only way new road construction can be financed in some places. And if the capacity is genuinely new, this can allay public suspicions that they’re being asked to “pay twice” for the same pavement.
But despite the enthusiasm of organizations like the Reason Foundation, the debate about whether to finance road construction using tolls or gas taxes has nothing in particular to do with freedom or free markets. I’m skeptical of any proposal that involves dramatically expanding the government’s ability to monitor and control the peoples’ movements—even if those new powers are deployed in the service of a “market-based” scheme of road pricing.
I’ve heard that some supermarkets have wanted to have express lanes for those willing to pay with pin debit. They are currently prohibited from doing that because of the merchant contracts Visa and Mastercard have put together regarding interchange fees – no discrimination allowed by plastic type.
Luckily if interchange reform survives rule-writing, that can happen, and people can forgo rewards points to get through the line faster and the supermarket will save money.
(Sorry, always hustling my pet projects.)
This is a great article, and I think this it contains a lot of insight. I’m thinking back to the South Fulton, Tennessee fire department that stood by and let a man’s house burn down because a $75 fee was not paid. This is a good example of what can go wrong when government adopts a so-called ‘free market’ policy. In a true market situation with viable competitors, any private fire company that did that would shut down by their next payroll. However, I wouldn’t be at all surprised if the city actually brought in more cash after the fiasco. (I’ll bet everyone in the area made *sure* their $75 was paid in full after all that!)
The idea that a political group has no incentive to properly set tolls to achieve maximum throughput is spot on, and we already have evidence of just such behavior in other areas. Consider red light cameras. Sold to the public as an enhancement to safety, in many communities they have merely become revenue enhancers. To paraphrase you, the revenue-maximizing light-timing schedule is not the same as the accident-minimizing schedule. (To minimize accidents, yellow lights should be lengthened. Some cities have lowered their yellow times to enhance revenue, with the direct cost of additional accidents.)
As a strong free market supporter, I admit I have been caught thinking that any move in that direction by government is good. Obviously, there is much more to consider.
Too many points to make here. I’ll try to write it up more completely later but, in brief:
– HOT lanes are more like your supermarket example than real congestion tolling. Congestion tolling is more like prices at the supermarket going up by 5% at peak hours.
– Privacy concerns are real, but privacy is norm-based anyway these days, so I see it more as a political issue than any real game-changer.
– Roads are a utility, and any private company charged with managing a tolling system should be closely watched and highly regulated.
– Solutions to collective action problems like congestion are often counter to what individuals will chose on their own; I am not sure it has anything to do with what ‘preferences economists think [people] ought to have’.
It doesn’t undermine your point by any means, but this system is used in several areas. For example, a lot of theme parks sell “fast passes” that allow people to wait in a shorter line, and first-class passengers get to board first in airplanes. I wonder exactly what factors lead an industry to adopt this practice or not?
Dan beat me to it about the theme parks, although I don’t think they actually “sell” fast passes at Disney Land – they give them away for certain time periods.
To be honest, I actually wouldn’t mind paying a couple of bucks to have a “fast lane” if I’m paying for a fairly large amount of groceries.
Congestion pricing is not a free market reform. Roads are a public, rather than a consumer, good. In fact, congestion pricing is being suggested not to fulfill some ideological ideal, but because it has actually been tried and has actually worked, most notably in London. There its main success has been to educate people about how much the maintenance of roads cost.
And you can’t possibly suggest that the English are less socialized to lines than Americans.
Differential peak and off-peak pricing for parking is another way that markets try to deal with time-of-day specific demand patterns.
Supermarkets have responded to check-out congestion in at least one way – express lanes for shoppers with a number of items below some threshold. The two most common thresholds I’ve seen are 10 and 15. This could reflect an attempt to separate customers by the size of the congestion externality that they impose on others. Or it could be in recognition of the fact that consumers with fewer items are more likely to simply walk away, which may create additional costs associated with reshelving and, for some items, loss.
This seems like a potentially rich area for study and analysis. The identification of social norms around queuing as part of that analysis seems important and may yield insights if applied across societies with different norms.
Tim, your piece in Ars was great.
I was going to make Dan Miller’s point about airlines having express lanes for first-class and elite passengers. I think to a large extent, the free market does provide something similar in grocery stores. Instead of first-class status at a particular store, people who don’t want to wait in line at the store patronize nicer stores.
Compare the checkout lines at your local Whole Foods compared to your local ghetto supermarket. I’d be willing to bet that there are more cashiers and shorter waits at the Whole Foods. People do trade off convenience for money in the free market.
Good discussion. There has been a lot of support for HOT lanes as a “market” solution, but this pretends as though this is a truly private enterprise. In Northern VA the 495 HOT lanes project was agreed to by local politicians with only a brief public discussion. The details of the contract that VA agreed to with Transurban-Fluor are sparse at best (my credit card contract has more pages than what has been shown to the public for this multi-billion dollar project). A few details of this truly awful agreement that have come out include guarantees by the state that Transurban will meet certain revenue and user number projections or VA will pay them the difference. They also used eminent-domain to seize land on both sides of the freeway, removing tree barriers going into neighborhoods and private businesses that are now abutted to the highway. The HOT lanes projects are like stadium giveaways but worse! In both cases the government pays for a private company’s investment, but in the case of HOT lanes the government pays for an investment that no sensible private company would have, and then guarantees it will be profitable. It’s corporatism at its worst. None of this is to say that private roads are bad or congestion pricing doesn’t work, but rather that the public-private partnerships like HOT lanes lead to projects that wouldn’t have and shouldn’t have come about without the government’s distortion.
Another form of congestion pricing in supermarkets: the self checkout scanners. These are NEVER busy at all of the supermarkets I frequent. The cost is that you have to scan and bag your own groceries.
Or you could always do this yourself and pay the person in front of you a dollar to take their spot. And then pay another dollar to the next person….
If the market wants to decide this, it can be done in much easier ways than having a bunch of idiots in a focus group poll people about the practice. If you want my spot in line, offer me a price and maybe I’ll take it.
Let’s say I’m heading home during the post-work rush and I’m in a hurry. All I need is a six-pack of beer and a quart of milk. I know it’s cheaper at MegaMarket but I can pop in and buy it quicker at the higher-priced ConveniaShop. I pay a premium to shop at the convenience store because I want to get home quicker. That sounds like a free market situation to me — and a lot better analogy to roadway congestion pricing.
Two other examples of off-peak pricing differentials: restaurant blue-plate specials, and movie matinees.
This is comparing apples to oranges.
I’m on the board of a cooperative grocery store startup, and given that we’re trying to get a $4 million store off the ground in a down economy in an industry where profit margins are fairly sparse, we spend a lot of time thinking about how to improve the bottom line over those critical first few years in business.
The difference is that the grocery store industry, more than almost all other forms of retail, is about volume. Food is such an enormous amount of people’s budgets that they shop hard for very small discounts, so your room to increase your markup is very low. So, the grocery retail industry is, at its heart, all about how to get as much food through the loading dock into the back, onto the shelves, and out the door as fast as possible with as little overhead as possible. Store managers already have every incentive to move you through quickly — revenues are almost 100% a product of the number of people you can get in the store, how fast you get them out, and how much they buy when they go.
Another fact about the grocery industry — your sales volumes are going to be heavily bunched towards just a few times a week, particularly Friday, Saturday, and Sunday. (You want to not stand in lines? Shop early on Tuesday afternoon.) Store managers will frequently throw all available personnel at checkout and bagging on those days, including multiple managers. Expanding beyond that means hiring additional workers, training them, and finding hours for them throughout the week, which adds to staffing, the single largest overhead expense at groceries. So, charging $6 to express lane customers doesn’t make sense, because all it does is require you to hire staffing, and probably doesn’t even make you as much money as just opening another normal lane and getting people out that way.
None of this, of course, applies to congestion tolling. In the grocery business, the biggest expense in the lanes is staff. In transportation, it’s all physical infrastructure construction and maintenance. Congestion tolling is primarily a way to raise money for additional infrastructure without reducing the extant value of the public good. In the grocery business, you can raise more money simply by increasing volume, but that effect is very indirect in transportation. Ideally, increased transportation capacity leads to greater economic activity which leads to greater tax revenue, but it’s far more indirect. Which, in a nutshell, is why “libertarian” or “free market” analysis of public goods like transportation capacity falls on its ass in a messy heap. (cleanup on aisle 3!) Transportation infrastructure is provided by the government as a means of improving economic activity; in fact, it’s probably the single greatest government economic development program ever invented. The true free market reform of roads is simply to put tolls for driving a car on every single one of them, right down to your neighborhood streets. This is, of course, insane, and would destroy the economy within a month, simply because not only would the complexity of collecting those fees be astronomical, but would add a major disincentive to economic activity everywhere. As such, the consensus means of financing transportation infrastructure is through taxes that tap general economic activity (income, property, and sales taxes), specific use taxes (car and gas taxes), and then top it off with supplemental income like congestion tolling.
Libertarians never like hearing that, for some reason, though…
I submit that grocery stores do congestion tolling now — by retagging and raising their prices on Saturdays and Sundays.
If they could raise prices from 4-7PM weekdays without causing a stir, I’m sure they would do it.
They also do it by price discriminating between fast service/full service stores (e.g. Bristol Market, Whole Foods) and slow service/partial service stores (e.g. Food4Less, HEB Pantry, Winn-Dixie).
In any case, sign me up. I’d gladly pay $0.10 per scannable item to get access to a shorter line.
I actually think that, with the right service design, there’s a viable business case behind optimizing the grocery checkout experience. Maybe it’s a per-item surcharge, maybe it’s a flat fee, maybe it’s a preferred membership subscription, or maybe it’s something radically different (and not-yet-invented) that allows customers to bag their groceries while they shop and then just wheel the cart out the door (sans “checkout”) as each item’s RFID tag, as well as something in the shopper’s smartphone, processes the transaction behind-the-scenes.
More importantly, I think the tricky part is predicting customer perceptions of the new business model. Without a tangible conception of the idea, it might seem like something consumers would never pay for. But show them a “magical” experience (or even just a couple of customers zipping to the front of the line on a Sunday morning before a Packers game in Wisconsin), and suddenly customers might appreciate the value (both emotional and rational) of paying extra to avoid congestion.
In the case of your customer support example, some companies have tiered support such that there is a basic level at no or low cost, and a quicker premium level at a higher cost. Checking Microsoft’s support pages, I basically have three choices if I want help with MS Word. I can try the support database or customer forums for free, I can pay $99 and get help in one business day by e-mail, or pay $259 and get help by phone within 4-8 hours.
Another example we will be increasingly dealing with is peak pricing of electricity through the use of smart meters. (For consumers. Large commercial users have been doing this for some time.)
The “wait your turn” norm isn’t so strong in other settings. Some nightclubs sell memberships that let you skip the queue. http://www.fabriclondon.com/fabricfirst/about
Disney has “express passes” for individual rides but they also sell some sort of dream ticket (I forget the name) that lets you cut all the lines.
There’s a fundamental difference between grocery stores and roads. Grocery stores are in the business of selling groceries. That is their primary good, and they are allocated based on prices. The checkout process is an ancillary service (which is expedited by self checkout and “X items or less [sic]” lanes.
By comparison, use of roads *is* the primary good. What else is the road selling? You are paying (or not) for the use of roads, and tolls can be used as a mechanism to calibrate supply and demand.
FWIW, there are a number of computer and software companies that have different service levels based on willingness to pay–free email support, pay for live support, pay even more for instant live support as opposed to a call back. Granted, it’s not dynamic pricing in the sense that it changes depending on the demand others are currently imposing on the support personnel.
Where things get hairy are the circumstances where public services are indeed involved. such as the example of the fire engines mentioned by Sean L. This is coming about where there is monopoly (or near monopoly) supply and no differentiation of service levels. For example, I am involved with a jazz club in Dalston in London. It has recently been “suggested” that we contribute to extra policing for the night time economy, along with other bars and clubs in the area. This is being described as part of the new Big Society initiative in the UK.
So far, no problem.
Except that the amount being sought is way beyond our ability to pay. We can’t raise our admission fees and pass it on to the customer. Does that mean that the police will boycott us when out at night? Or should we start employing freelance vigilante groups? Or should we just close down – a hard decision to make because of the investment made in terms of time and money over the past years?
Congestion pricing will become more prevalent in the private market over time, as consumers adjust to it.
The most prevalent example is congestion pricing by cable companies, ISPs and telecoms where users are charged different rates for higher/lower rates of data transfer.
Like the Disney example cited above, there are paid express lines at certain ski areas. At Northstar in California, skiers can buy a “Fast Pass” for $20/day (on top of the regular ticket price), which allows one to avoid long waits of up to 45 minutes.
As for your supermarket example, many stores offer online shopping with in-store pick up for a charge.
I think you’re right when you say that “customers would see people using a tolled checkout lane as breaking an implicit social contract.” When I bought the Fast Pass at Northstar, I actually felt guilty. But I skied a lot more on that particular day than I would have otherwise (I think there were 10,000 skiers on the mountain that day). I will probably buy the Fast Pass again. All of which is to say that markets have a perverse way of altering mores.
Supermarkets actually already do this with self-checkout lanes. These lanes offer customers willing to “pay” more in the form of doing labor that a supermarket employee would normally do a quicker checkout experience. I don’t resent people that use the self-checkout lane. And I use them myself when I am in a hurry.
Perhaps then the solution is finding more congestion-skipping services that operate by reducing the cost to the service provider rather than increasing the cost to the consumer.
Carpool lanes are, kind of, an example of this. In the nightclub example, perhaps we could offer express entry to those who buy entry to the club as well as 3 drink tickets. Restaurants could offer express seating to those who commit to being at their table for a maximum of 1 hour. Non-guaranteed tickets (to anything) operate similarly.
Its not just congestion charging but bus lanes in the UK that cause an affront to people’s perception of fairness as this item notes
Bus Lanes and a British sense of fairness
The national debate resulting from the Government’s decision to remove the bus, taxi and motorcycle lane at the London end of M4 can be seen amongst other things as, a classic indicator of the London bias of the national media (why should such a local parochial issue be making national news) or an issue that goes to the very heart of the British notion of fairness.
It is often claimed that the British love their queues, we wait patiently in line at the post office, the supermarket checkout, and still in some locations at the bus stop. We dislike the queue jumper. I still vividly remember the boos and catcalls at Moscow Sheremetyevo airport over 20 years ago as the Brits arriving on a BA flight diligently queued up at passport controls while the Russians and other more regular travellers followed the local custom of the rugby scrum approach to get to the front of the queue.
Closer to home it is enlightening listening to the comments of fellow queuers at the nation’s amusement parks as people who have paid for the various priority queuing schemes jump to the front of the queue. They are generally uncomplimentary and the most used phrase is that “its unfair” and that’s from the adults.
There is still a deep seated notion of fairness in the nation, a view that we should wait our turn and that no one else should be able to jump the queue.
This attitude spills over with regard to bus lanes. The car driver (if you assume the main motoring organisations speak for the majority) believes that it is unfair that they should be excluded from certain road space. Especially when that road space does not seem to be well used. They not surprisingly believe that removing road space and letting other users jump ahead of them results in more congestion and delay for them.
In part this is due to the failure of traffic engineers and transport economists to adequately explain the fact that bus lanes actually have minimal impact on journey times for car drivers (its junction capacity not lane capacity that dictates the level of traffic a network can accommodate), have significant advantages for bus users and so overall provide substantial benefits to the community as a whole. With the advent of micro-simulation it is far more easy to show the with and without bus lane situation and the professionals need to be more proactive in the use of this technology to demonstrate more effectively the advantages of bus lanes and the fact they have minimal impact on general road users.
But for even those who you might expect do understand these impacts of bus lanes the response to the abolition of the M4 bus lane is disappointing and seems to reflect this notion of fairness rather than efficiency. It was perhaps not surprising that the Secretary State used the word unfair in his announcement while tagging on the economy at the end “my intention is to scrap the bus lane permanently as I believe excluding road users in this way is unfair and damaging to the economy.” The AA welcomed the abolition of the M4 bus lane even though admitting that it would make no difference to journey time.
“We are very pleased to see this development, as we have campaigned against the bus lane from the very beginning. It is always aggravating to sit in a traffic jam beside a bus lane that has nothing in it, and that is the situation on the M4. The only people who will mourn it will be London taxi drivers, who have been able to use it whether laden our unladen,” said AA public affairs spokesman Andrew Howard.
But he added “When the bus lane ends, the M4 then reduces to two lanes for an elevated section. So you are still going to see big queues going into London – getting rid of the bus lane will simply move them 3.5 miles to the east.”
So when I asked Edmund King the AA President (via Twitter) why given Mr Howard’s acknowledgement that the abolition of the M4 bus lane will have no impact on congestion the AA still supported its removal his response was “for M4 lane to be effective needs buses, park and ride, car sharers otherwise white elephant”. The paradox of course is that the busier the bus lane the greater the level of delay for other users.
It appears when it comes to queue jumping the British don’t like it whether its buses on the M4 or priority boarders at Lego Land.
I actually propose we like a certain amount of congestion. My, as yet unpublished but being written up when I get round to it, research suggests OK people don’t love congestion but they prefer it to any demand management intervention and here’s some suggestions as to why:-
It is often viewed as natural and organic – congestion happens as a result of people needing to be there – no one cause, lots of little ones; demand management is an artificial and distant proposition singular intervention that interfers with the naturalness of the jam.
Dunkirk spirit – people love to talk about it – they way they had to battle through and win in order to get to their work. Everyone was in it together and managed to pull through. Great source of story telling and is an acceptable excuse for being late for things
So, is congestion a psychological need?!
Neanderthal Man approach – Tough and hard –beaten through congestion to get to work/home – a sense of achievement for getting through it
Identity – I’m that important that I need to sit in that much congestion every day to do what I need to do.
Source of story telling, discussion and boasting
Protected space and time – free from role responsibility (the gift of travel time)
Social yet distance – amongst others you don’t have to touch, smell or talk to. The need to be part of a “crowd” without having to interact directly.
Protected bubble – free to sing, smoke, shout and be yourself – where else can you do that these days?
Normality and habit – Ordinariness of being in a jam – its expected.
Never know any different – Younger drivers
Years ago, one of the contributors to Journal of Economic Perspectives related a story about his attempt to buy the next number to be served at the butcher shop, in order to get the steaks home and on the grill more quickly. He left without any steaks after making everybody else in the queue angry at his departure from the norm.
One of the challenges to modelling the implementation of congestion pricing is capturing the possible loss of goodwill. The amusement parks that sell priority seating apparently expect the consumers delayed by the policy to shrug it off; sometimes the same priority is offered to ambulatory guests, which cloaks it as accommodation of a disability.
Perhaps the separating equilibrium with the supermarkets is different, in the way that several commenters have suggested: people with higher opportunity cost of time patronize tonier outlets, on average pay more, but on average get out more quickly, compared to the discounters with three cash registers of ten open at any time; or they fill their shorter shopping list at a convenience outlet and pay a higher price, but unless everybody else in the queue is buying lottery tickets and Marlboros, they get out more quickly.
I think you are confusing “consumer preference” with the predictable result of government policy.
First of all, you do not mention alternatives to driving once in this post. In the only place that congestion pricing has actually been considered, New York City, more than half of the people don’t use cars for their daily commute. I suspect this is because New York has built an extensive and convenient public transportation system that offers consumers the “choice” of not driving.
True this choice is not offered in many cities, but I think most of the support for congestion pricing from transportation experts is paired with the idea of providing the choice of riding public transportation. Public transport is seen as beneficial by many people because of the lower externalities associated with public transport vs. driving such as emissions and parking space.
Also, your perception of Elites vs. The People is a joke to those of us in big cities. Who do you think still drives to work in New York, San Francisco, DC every day? It’s not the millionaires riding the bus.