Economics for Dummies
- Posted by ryan on November 15th, 2008 filed in Economics, In the News
I was going to write something big on this post at the Oil Drum, arguing that suburbs would be ok in an age of peak oil. Now I’m just going to say two things about it. One is that “suburbs,” in this context, is all but meaningless. In the Washington area, “suburbs,” includes Brookland, Bethesda, Ballston, Potomac, Tysons, Woodbridge, and Winchester–wildly different looking places. Peak oil, or high fuel prices, will have dramatically different effects across those locations. Also, he leaves out congestion. You can’t have a discussion about the fixed and variable costs of commuting and optimal decision making without discussing congestion.
I’m just going to say two things about that post, because in preparing to write something big about it, I came across this post, the first in the author’s suburbs series, on sunk costs. He writes:
Sunk cost is the economic concept that some costs, if they cannot be recovered once they have been incurred, have significant effects on our decision making. What is the sunk cost of Suburbia? Individual homes, for individual buyers, may not entirely represent “sunk cost” if they sell immediately, though the decline in prices over the past months does represent sunk cost. If everyone in suburbia wanted to leave, however, then the entire suburban project–tens of trillions of dollars–would represent a sunk cost.
In layman’s terms, if you bought your house for $200,000 but can only sell it today for $50,000, then your sunk cost is $150,000. Even if you didn’t have a mortgage, that would represent a significant disincentive to selling.
Nope! Big whiff. He’s right that a sunk cost is the unrecoverable share of a past expenditure. He’s very, very wrong that a big sunk cost represents “a significant disincentive to selling.” That’s the sunk cost fallacy. We feel like it should be a disincentive, but it isn’t. A sunk cost is utterly irrelevant to any new decisions.
A classic sunk cost example is the movie ticket. Say you spend $12, or whatever they’re charging these days, to see Gigli. You get in the theater and quickly realize you’ve blown twelve lottery tickets on absolute garbage. You want to leave. You’d be better off leaving. If you stay, you not only waste the time you spend in the theater, you also get the negative utility that comes from watching Jennifter Lopez trying to act. Any rational person would leave, and fast.
But, you say, I dropped $12 on this. I really should get my money’s worth. But you already made the decision that ensured you wouldn’t get your money’s worth! Nothing you can do will get you that $12 back. It’s gone now, irrelevant. You can either be miserable for 90 minutes or do something productive with your time. And so, a smart person leaves the theater.
Ditto with somebody who owns a $50,000 suburban home. The difference between what you paid and what it’s now worth is gone, and it isn’t coming back. The only thing relevant now is the payoff to different decisions. Whether you own the house outright or still owe a lot on the mortgage, the choice is straightforward–do you get more utility from leaving than from staying? At no point does the vanished $150,000 enter into the discussion.
Even an outstanding loan balance should not affect the decision. You owe that whether you stay in the home or not. It’s not going away. It may add new variables to the decision making process (should I default, or sell and keep paying?), but because that $150,000 is unrecoverable, you have to forget about it. You may as well have racked up the debt on your credit card gambling in Vegas. It’s gone, baby, gone.
This is actually a really important point for economists to get across. You have folks that own homes in towns where the local economy has been dead for decades. Given a big enough emotional attachment, it could possibly be in their interest to stay, but for most such homeowners, they’d be better off taking what they can get for their homes and decamping to more vigorous markets. But a lot of folks will remember that they paid twice as much for their homes as those homes are now worth, and so they need to live in them to get their money’s worth. No, no, no! That value isn’t coming back! It’s not a disincentive to sell; it’s irrelevant.
There may, in fact, be some interesting, behavioral economics oriented policy solutions out there that take into account the sunk cost fallacy in addressing the mortgage crisis and the problem of immobile households.
November 15th, 2008 at 4:59 pm
I’m not sure that, in the actual situation that’s being discussed, it’s actually a sunk cost that’s the disincentive.
The situation arises now in upstate New York. You have dying towns, dying cities. The young leave; the middle-aged and old tend not to. They bought houses there, perhaps 20, 30 years ago. They now own the house free and clear, or with a small house payment, set, after all, twenty-odd years ago. On the small income they can earn staying in the dying town, they can keep a roof over their head. But they can’t sell the house for enough money to allow them to buy another elsewhere, even given that they could earn more there. It’s not the sunk cost of the house that keeps them immobile; it’s the difference between house prices where they are and house prices where there are opportunities. I imagine a similar dynamic holds in part of the rustbelt, too. I don’t know how to label it: I think it’s some variant of opportunity cost.
The suggestion is that what’s now true of the North Country may well become true of (some parts of) the suburbs. That people who now own houses in, say, Prince William County will, as transportation costs inexorably rise, find themselves stuck. They’ll be able to keep making their current house payments, as long as they can find work locally, but they won’t be able to sell their houses and buy another in a more transportationally desirable neighbourhood.
I don’t know if this sort of scenario will occur, or is even likely, but don’t dismiss it because the guy mislabels the problem as one of sunk cost.
November 15th, 2008 at 5:43 pm
It may also help to know the context of the argument, which in this case is the Kunstler Collapse of Suburbia thesis.
In that context, there is an argument involving sunk costs that Kunstler overstates his case. That is, Kunstler’s argument is that modern suburbia (by which he seems to mean outer suburbia and exurbia) has too high an energy cost to be sustained as the effects of Peak Oil kick in.
However, there is a difference between new suburban development and existing suburban housing … even if we reach a point where sprawl development is no longer viable due to its energy intensity, that does not automatically mean a “collapse of suburbia”, because some of those energy costs are sunk costs … for existing housing, they have already been expended.
Given those sunk costs, then even under the Kunstler’s presmises on Peak Oil and high energy costs, a long slow decline seems more likely than Kunstler would suggest, and a sudden collapse less likely. So, while the author of the post mishandles the concept of sunk costs, sunk costs are a relevant concept for consideration of the Collapse of Suburbia thesis.
November 15th, 2008 at 7:11 pm
The Collapse of Suburbia thesis is wishful thinking by a famously anti-Suburbia writer who is widely regarded as a kook.
As Ed Glaeser among others has pointed out, suburban commuters can offset rising gas prices by the simply expedient of switching to a more fuel-efficient vehicle. Trade-in that SUV for a Corolla or a Prius and you’ve cut your fuel consumption in half or by two-thirds. In the future, many more hybrid models will be available, as well as plug-in hybrids and fully-electric vehicles that will be even more efficient. There are of course also other ways of offsetting rising gas prices, such as telecommuting and carpooling.
As for congestion, it is worse in high-density areas than in low-density ones.
November 15th, 2008 at 11:23 pm
“There may, in fact, be some interesting, behavioral economics oriented policy solutions out there that take into account the sunk cost fallacy in addressing the mortgage crisis and the problem of immobile households.”
The whole-herd buyout program in the 80s, meant to reduce a milk glut and isolate unprofitable dairy farmers to save the more efficient ones might be an example of this. I was milking four cows at the time and a little hurt we weren’t contacted.
November 16th, 2008 at 2:43 pm
If this argument is correct, shouldn’t it also apply to losses in the market? And shouldn’t the wise thing to do be simply to walk away rather than continue investing for the future long term of indeterminate length? Aren’t these losses sunk costs? If so, why are (at least some) economists still telling people that the market is a good place for people’s money and that now is the time to buy because stocks are now a bargain? I’m confused. If something is an economic “truth,” “law,” shouldn’t it apply universally rather than selectively? If not, why not? I don’t know whether the real title of this piece is “Economics for Dummies” or “Dummies in Economics.” Help me, please!
November 16th, 2008 at 2:55 pm
John, the situation is the same if you, for instance, purchased a tech stock at the height of the tech bubble. If its price collapsed and there is no expectation of its ever recovering its value, then that lost value is a sunk cost.
What economists are saying about stock markets now, for the most part, is that fear has excessively depressed prices. Based on historical price/earnings ratios and annual earnings growth, stocks are a good buy. So if you’ve lost a lot of money in the market, the question is just where do I get the best return, looking forward. That very well could be the stock market.
November 16th, 2008 at 4:06 pm
But while “sunk cost fallacy” is the rational way to look at things - who said people are rational?
Or at least isn’t it fair to expect that many of them will take a while to face up to the fact that various costs are “sunk”?
November 16th, 2008 at 5:23 pm
I think there’s another dimension of the “collapse of suburbia” - better collapse of exurbia - thesis. Housing prices are affected by the reputation of a neighborhood as well as tangible factors.
As the house prices in outer suburbs go down, living on big lots will become less prestigious and there will be a self-reinforcing cycle, the inverse of what happened to inner cities in the fifties and sixties. Many of the outer suburbs may become cheap housing for people who drive to park-and-rides and take the bus to work.
November 17th, 2008 at 10:16 am
Kunstler isn’t “widely” regarded as a “kook”. He’s out there, but has interesting and valid insights as well. Hyperbolic is a better descriptor.
Nonetheless, I think the ‘burbs will not see a precipitous decline, as all that investment in money and dreams cannot collapse - this will alter society in fundamental ways.
In addition, not everyone wants to live in density, so someone will have to figger out how to have these far-flung neighborhoods hang on.
The latest JAPA (pp 419-480) explores the shrinking city and the hanging on part, with a series of very interesting papers.
Certainly folks like their ‘burbs. But not everyone will stay there. What to do about the population shift?
November 17th, 2008 at 12:03 pm
A couple of years ago, I was at an MLA panel on the rhetoric of peak oil. One of the papers was on Kunstler. The presenter, not sure how much his audience knew of Kunstler and wanting to describe him succinctly said, “Think of Jonathan Edwards with a weblog:” suburb-builders in the hands of angry peak oil.
He overstates for effect and, particularly in the novels, telescopes timelines. He needs to be argued with, though, not simply dismissed. One of his arguments for rapid suburban decline is that they’re jerry-built. If undermaintained for any length of time, the suburban housing stock will rapidly deteriorate, unlike the legacy durable 19th and early 20th century housing stock in American cities, which has endured 50-60 years of undermaintainance and remains recoverable.
November 17th, 2008 at 12:05 pm
They only think they like their ‘burbs because they don’t know any different. They were never given a choice and generally don’t know what life in a normal, non abandoned walkable place is like. They have no means of comparison.
Talk about a sunk cost. Good riddance to suburbia. If people like their their faux-country living, they should pay for it rather than having the city subisidize them.
November 17th, 2008 at 5:52 pm
I hate to belabor this, but I can’t make much sense of this. I believe you’re telling me that a sunk cost is one for which there is no realistic expectation of recovery. But if there is an “expectation” of recovery, the cost is not really “sunk.” So write off the house but invest in the market. The action to be taken is not based on the money lost, which is irrelevant, but on some “expectation.” If this is correct, I believe it’s unworkable. Expectations are wishy-washy things. A lot of people had strong expectations that they could save for retirement by investing. Their expectations were erroneous. Someone believes that a buyer for his house will come by who will see something in it that will cause him to pay more for it than it’s current estimated value. His expectation may be right or wrong. Although the money lost isn’t what keeps him from selling, it certainly plays a role in creating his expectations. Isn’t this situation true for almost any situation? There are, after all, people who still expect the second coming. Is their expectation unrealistic? If the determination of whether or not something is a sunk cost depends on an expectation, it is a useless concept.
Even the theater ticket example baffles me. It’s far too simplistic. I can conjure up a thousand reasons why a person who discovers that he was rooked into buying tickets to a bad movie would choose to stay and see it out, all the while acknowledging that he has wasted the money. Perhaps even the obvious one that he has nothing better to do at the moment anyway. So if there is no way to really determine when something is a sunk cost, how does the concept help one make decisions? And if it doesn’t, phooie! There’s something seriously wrong with all of this. I’d like to know what it is.
November 17th, 2008 at 6:08 pm
Ok, with the theater example. Think about it like this. If you have a good reason to stay and watch the movie, then stay and watch the movie. But the $12 you spent on the ticket isn’t a good reason, because it’s gone whether you stay or leave. Something is a sunk cost, and therefore irrelevant, once it becomes unrecoverable. That doesn’t mean you have to make one decision or another, it just means that you shouldn’t take into account the sunk cost.
November 17th, 2008 at 9:11 pm
Doesn’t make sense to me either. If the movie is worth anything to me at all, why am I not better off staying to watch it rather than leaving, given that my $12 ticket is a sunk (unrecoverable) cost? If the actual value of the movie to me is, say, $6, then by staying to watch it my loss is only $6. But if I just leave without watching it, my loss is the full $12 cost of the ticket. So the economically rational choice is to stay and watch the movie.
Conversely, if the ticket is not a sunk cost but is refundable, the economically rational choice would be to leave without watching the movie and collect the refund. That way, my loss would be $0.
November 18th, 2008 at 12:05 am
Mixner, think in terms of opportunity cost. Let’s say that staying to watch the rest of the bad movie is worth $6 to you. But let’s also suppose that spending the next two hours in a bar watching a football game is worth $10 to you. Then you are $4 better off if you leave. Or, put differently, your loss is only $2 if you go watch the football game.
You minimize your “loss” if you do the thing that yields the highest return. How much you spent on the ticket ought to be irrelevant to that decision.
November 18th, 2008 at 3:28 am
AC,
You minimize your “loss” if you do the thing that yields the highest return.
Right. But that thing may be staying to watch the movie. Unless your benefit from doing something else would be more than $6 (or whatever watching the movie is worth to you), you’d get the highest return (smallest loss) from watching the movie. But only because you can’t get back the $12 you already spent on the movie ticket. If the $12 were recoverable, the equation would be different. So how is the sunk cost of the movie ticket not relevant to your decision?
November 18th, 2008 at 4:23 am
Sunk cost is irrelevant once it’s sunk; all that matters is what options provide the best return assuming the initial investment is not recoverable.
Now if the sunk cost is not really sunk, that does change the calculus. But then we’re not talking about the sunk cost fallacy.
In your hypothetical: You pay $12 for bad movie; it’s worth $6 to sit through it; you’ve got nothing else worth doing for $6 so you stay. But you find out you can get a refund — the cost is not really sunk. You get your money back and save a $6 loss.
November 18th, 2008 at 1:50 pm
We seem to be talking about different things. Yes, given that a cost is sunk, the size of that cost isn’t relevant to subsequent choices. But whether the cost is sunk or recoverable definitely is relevant.
November 18th, 2008 at 2:25 pm
Ok. I give up. If I had chosen to keep this going, my next questions would have been, What constitutes a good reason? And why is it impossible for a good reason to have been influenced by the amount of money lost, in which case the amount of money lost is not irrelevant?
November 24th, 2008 at 12:04 pm
I think you go too far when you say, “A sunk cost is utterly irrelevant to any new decisions.” That’s not accurate at all.