(note: if you wish to clear the history for past days, hit the "refresh" button on your browser first; otherwise, clicking this button will add to previous totals)

Current time:

The parking lot

Customers arrived successfully

Alice0
Bob0
Charlie0

Customers lost due to lack of parking

Alice0
Bob0
Charlie0
Businesses allow sharing
Bob and Charlie only allow sharing

Tragedy of the Commons? Or Annoyance of the Parking?

Economic metaphors as a shortcut for thinking

Many of us are aware of the phrase “Tragedy of the Commons”, without ever having read the original essay by Garrett Hardin. In brief, it uses the metaphor of a commonly owned pasture to illustrate the problem of a resource that anyone can use, but no one feels particularly responsible for. If every herder can graze his cattle (or whatever) on the commonly owned pasture, then there will tend to be a problem of overgrazing. Even if every person acts rationally, with an awareness of the potential problem of overgrazing, the end result may be that the commons are nonetheless overgrazed. So goes the story, at any rate, and most of us can think of examples of many people acting in a short-sighted manner, in part because they don't trust others to think of long-term consequences and therefore have no incentive to do so themselves. If I know the commons are destined to be overgrazed, what reason to not graze my kine while it's still possible?

Leaving aside the question of whether events actually play out this way in the case of grazing on common land (the short answer is “maybe, sometimes”), there is the larger question of the usefulness of the commons as a metaphor. Every metaphor has limits in its applicability, but the darned things are so useful as a mental shortcut that it is easy to over-apply them. It is handy to have a countervailing metaphor available. Do we have one that is commonly understood, that we can use? I believe that we do.

Parking, in many American cities, is an annoyance. There are some urban areas where the population density is so great, that any car-based system will break down, owing to the fact that (relative to streetcars, buses, taxicabs, etc.) the privately owned automobile spends far more time parked and empty than it does conveying people from place to place. Moreover, it is often parked in a crowded part of the city, where space is at a premium. In these cities, no policy will keep parking from being problematic; it will be either horribly expensive or horribly scarce, or both.

There are on the other hand areas which are far below this level of population density, and automobile transport is not a problem there. Parking is readily available almost anywhere that people want to go, and rarely the subject of much controversy. The interesting case, for our purposes, is when the population density is between these two extremes, as is often the case with small cities or large towns.

Consider three businesses, Alice's DryCleaner, Bob's Diner, and Charlie's Coffeehouse. Let's say that Alice gets most of her business between 7 and 8 am, and between 5 and 7 pm, and a typical visit is for 15 minutes. Bob gets most of his business between 12 and 1 pm, and they stay for an hour and 15 minutes. Charlie gets most of his business between 6 and 9 pm, and they stay for an hour and a half each.

Crucially, these are just typical times. Some customers drop off at Alice's in the evening, some go for dinner at Bob's Diner in the morning. But they each have peak times. Another crucial factor is that their customers have other options; if Alice's is too busy, they will go to a rival dry cleaner, and the same for Bob and Charlie's customers. Thus, a lack of parking at any point in time leads directly to lost patronage.

Alice, Bob, and Charlie each have parking spots available for their establishment. They have several choices in how they regulate that parking. Most obviously, they could require that it be used only by their customers, and only while those customers are patronizing their establishment. They could also pool their parking, probably still excluding anyone else, but allowing any customer of any of the three businesses to park in any of their parking spots.

To simulate this, and see which strategy is preferable, we will write a program with the following rules:

  1. half of each business' customers come during peak times, half at all other times combined
  2. Alice owns 1 parking spot, Bob owns 2, and Charlie owns 3
  3. would-be customers who arrive and find no parking available will leave, and that business is lost forever (i.e. they don't just defer that trip, but instead go someplace else)
  4. enforcement of parking regulations is absolute, either because of zealous towing or because fear of towing prevents anyone from challenging it

For our simulation we have 60 potential customers show up each day, 20 for each one of the three businesses. We ran the simulation 20 times, to simulate 20 separate days, and averaged the results.

First, the results when businesses do not share parking spots. You can simulate a day in this way yourself by clicking on the button on the left (leaving all checkboxes at the bottom of the page unchecked). Note that because there is some randomness built in to when customers arrive, your results may not be exactly the same as mine.

Second, the results when businesses pool their parking spaces. You can simulate this situation on the left by checking the box for "businesses allow sharing" and then clicking the button. If you want to reset the totals first, reload the page using your browser's "Reload" button.

Lastly, the results when two of the businesses pool their parking spaces, but one does not.

At the risk of belaboring the obvious, it is clear that pooling or sharing their parking spaces will result in a better outcome for all three businesses, even though the results are more dramatic for some businesses than others. Even when only two businesses share, those two are better off, and the third is worse off. There is some randomness in the results, so if you run them yourself you may get slightly different results, but if you run it ten times or more you should get something close to the results shown here.

There are, for any good economic analysis, two parts. The first is an analysis of the result for any given allocation of resources. The second part is an analysis of the psychology of the people involved. It is true that some economists dispense of the second part by claiming that people are perfectly rational and perfectly well informed, or behave economically as if they were, but this is more wishful thinking (on the economist's part) than justified by the facts. Thus, we should consider what real people who are in the analogous position to Alice, Bob, and Charlie actually do.

In practice, they do not have access to the simulation we have performed, or at least are not aware that such analysis is available (most business owners are not computer programmers). Those who choose to share parking, do so for one of several reasons (not wanting to antagonize neighbors, not wanting to antagonize potential future customers, not aware of the importance of parking). There are also a few cases where there is a calculated decision to share parking.

The first is when the parking is literally held in common, perhaps by a third party such as a mall owner. The business owners lease retail space in part because they believe that the availability of parking, and the traffic coming to that location to go to other stores, will benefit their own businesses. Another case is when two or more businesses are so similar in their typical customer that they are easily able to perceive a potential synergy between their businesses; customers who go to one business will often go to the other as part of the same trip, so why make them move their car a few spots down. For example, a book shop and a record shop which happen to cater to more or less the same crowd.

Failing one of these cases, though, business owners tend to jealously guard their parking spots, and in some cases contract third parties to enforce this. Consider for a moment the incentives for such a third party (towing company). They are paid by the (unwilling) customer whose car was towed. They have little reason to expect that they will ever have a satisfied customer. They are not expecting repeat business (it is more or less their purpose to discourage it). They do not want to antagonize their customers to the point of provoking legal action, but the optimum strategy will be to tow so zealously that they approach this limit as closely as possible. By design, they will annoy people.

Why, then would business owners tend towards a policy which will annoy their customers? Especially when there are examples available (e.g. malls) which demonstrate that a sharing policy can work?

The answer lies in a limitation to the feedback loop from customer to business owner. If a business owner sees a shortage of parking, and knows (from looking around their shop) that it's not because they have too many customers in their own store, they will perceive a (potential) loss of business. But, if they are instead profiting from the availability of spaces in other businesses' spaces, they are unlikely to notice the benefit. For one thing, they are busy dealing with a rush of customers, and are unlikely to look outside their shop windows to notice where people are parked. For another thing, perhaps more importantly, there isn't a problem for them to look around to find the source of. When they have few customers they will be anxious to discover why not, and if there is a shortage of parking they will blame their lack of custom on that (whether it is true or not, but let's assume for the sake of argument that it is in that particular case).

This behavior is, at the larger scale, irrational. In many cases, all businesses would be better off (over the course of a week) if all parking were shared. This simulation does not at all demonstrate that there are NO cases where a business would lose by pooled parking, but it is not hard to show that they often benefit, and the more businesses pool their parking in this way, the larger the likely benefit (since they are unlikely to all need the parking at the exact same time if there are many different kinds of business). But if businesses who pool parking benefit and businesses who tow do not, then why does the competitive pressure of the marketplace not squeeze out such shortsighted behavior?

The reason is, that business owners' behavior is not randomly determined by a host of independent personality traits. There is no “sharing parking” trait which business owners either have or do not have. What people do have (or fail to have) is a tendency to resent (and act on) loss of profit from the use of their resources by another. Most businesses have a shortage of resources, and owners who do not watch out for the loss of them will suffer from shoplifting and other types of behavior where they are “taken advantage of”; reacting strongly to this is a characteristic beneficial to business owners in general, even if it works against them in the case of parking specifically. It is similar to the phenomenon in which the gene that causes sickle cell anemia persists, even though it can cause ¼ of the person's offspring to sicken and die (if both parents have one copy of the gene), in those regions where malaria is an even bigger threat. Sickle cell anemia is itself bad, but because it can help avoid an even more costly threat, it can be a net benefit to have one copy of the gene.

Since business owners will tend to notice cases where they lose custom due to their parking being used by other businesses customers, and not notice cases where they benefit from the converse, they will all tend to conclude that they would be better off restricting their parking spaces to only their customers, even in cases where they all would be better off pooling their parking capacity!

This behavior may, in some sense, even be “rational”, in the sense that based on the cases they witness/notice, they are losing more custom from this policy than they gain. Unless the business owner is just as keen to notice when they are benefiting from pooled parking as when they are suffering from it (unlikely, for the reasons mentioned above), they will draw an erroneous conclusion regarding whether or not they are profiting from pooling their parking with their neighbors. When they are busy, they do not ask themselves “where are all these people parking?”, because they are busy. When they are not busy, they will look for something (or someone) to blame, and act to correct the situation (e.g. put up signs saying “parking for my customers only or I will tow”).

It is furthermore important to note that, the more inclined the store owner is to notice when they are having resources taken away, and the more inclined they are to inquire aggressively into the cause of times when they have few customers, the more likely they are to reach this erroneous conclusion. In other words, the same disposition which can (in most things) make them a successful shop owner can make them less likely to think clearly about this case.

There are, of course, business whose typical peak times are pathologically mismatched, as for example when one business has a peak time typically just before another, and whose customers stay long, as for example if a restaurant fills up with dinner customers just before the peak time for a movie rental shop. These cases are in any case a minority of the real-world situations where businesses choose NOT to pool their parking. Even when all business owners are generally good at their job, they are likely to divide up and jealously guard their own parking capacity, regardless of the fact that it would benefit all of them to pool their parking instead.

Therefore, I present to you “The Annoyance of Parking”, in the hope that it may function for you as an antidote to the over-application of “The Tragedy of the Commons”. To be clear, both can be apt in different situations, and we should not replace one over-applied metaphor with another (unlikely in any case, since the tragedy of the commons is so well-known that it will likely never be displaced in people's minds). But, if we understand that a potentially shared resource could be subject to either kind of problem, we are in a better position to make sure we pick the right metaphor to use to understand the situation at hand.

So, when you are looking at a resource which could be either held in common or held individually, ask yourselves, is it more like a grazing commons (depleted by use and only slowly replenished) or more like parking (usable by different people at different times)?