The country of Bubblia has a problem. It's a fairly large country, and there are 100 million cars in operation. The cars are essential to this economy, and if the cars don't work, the country will experience economic disaster.
For the last 30 years, Edsel, Inc. and all the other car makers in the country have been reliable. When any one of them sold you a car, you knew there was little risk of it breaking down. Some did break down, of course, but not very many. The default rate on cars was very low. Unfortunately, however, the companies recently came out with new models boasting the latest technology. These were based upon fancy mathematical modeling, and the promise was that they would be cheaper and more reliable than any other car ever.
But it turned out they weren't reliable at all, they are breaking down in massive numbers, and here's the thing. It's impossible to know which cars are lemons and which aren't. You'll be driving one of them, and one day, all of a sudden, it just blows up. When that happens, the car is basically totaled.
Every one of the 100 million cars could blow up at any time, and nobody knows when or which car it might be. The actual percentage of cars that have problems is not known, and there's no way to tell from inspection if the car is bad or good. And it's been a huge problem. Every day, in every major city several cars break down per hour on the freeways, streets, etc. and it completely clogs up transportation - this is a country that relies upon cars for its economy to function and it's causing economic distress. It cannot have frozen roads.
Alan Bernson, a loyal government servant in charge of the Federal Car Reserve, has an idea. We'll call it the "Bernson Plan". Currently, because of all the lemons in the car market, and all the uncertainty, cars are estimated to be worth $4,000 each, on average, though nobody knows for sure since hardly any are trading right now. But, if the estimate is correct, it would cost 400 billion dollars to purchase them all.
The initial plan is to buy up the all the cars and give people $4,000, on average, for them. The government will then spend six months, or longer as needed, determining which cars are bad and which are good (through driving them or some other procedure). After the test period is complete, the government will then take the cars that are just fine - however many that might be - and sell them back to the public at market prices. The hope is that once the bad cars are out of the mix, the price of the cars will rise above the $4,000, and in fact rise enough so that even with a fraction blowing up, they still make money. But so long as prices rise and the number of cars that blow up isn't too large, and they don't think it will be, the government shouldn't lose too much in any case.
But before the plan is put into place, Brad DeKrugen and others point out (in their best Swedish accents) that this won't work. There are losses, cars that no longer work. People need cars, that's what makes the economy tick, and $4,000 isn't enough to buy a new car. People need to be recapitalized for this to work - the cheapest new cars are $7,000 so people need at least that much to replace the old cars. If we only give them $4,000, they aren't going to be able to survive in most cases. Without a car, they go out of business, can't get to work, etc. Sure, the roads will be unfrozen, but not in the way we want as it will be because there are no cars on the roads at all, or very few. In fact, people are unlikely to even sell their cars in this case - better to keep the car and take a chance that it won't blow up than to have $4,000 and no way to get a car at all. And if they decide to keep their cars, the problem with the clogged roads persists until, after a long drawn out period of clogged roads and a clogged economy, perhaps a period of years, enough of the cars die on their own to solve the problem.
So Bernson is forced to reevaluate the plan. Dang, he thinks, I should have asked for way more than I needed up front. I'm going to have to offer $7,000 for these cars even though they are only worth $4,000. So I need $700 billion, not the $400 billion I have approval to spend, to get the job done.
So, Bernson goes back to the legislature, and, with a little bit of salesmanship (if you don't give me $700 billion, we may have to declare martial law!), the legislature reluctantly agrees. Their agreement is helped by the fact that members of congress have many constituents who are in the car industry (as cars are so essential to the economy). But the big push comes from car makers. Though they created the problem by producing lousy cars, car makers stand to benefit greatly from selling new cars to replace the old, so they lobby hard - and successfully - for the bill. The lame duck president, who seems not to care much one way or the other, signs the bill between bike rides.
So the trades finally begin. You show up at the designated agency in your town, give up title to your car, and in return get a certificate valued at $7,000 that can be used as a down payment on the purchase of a new car (and is enough to buy the cheapest model outright so everyone can get some type of car).
As for the government, it holds onto the 100 million cars it bought for six months during which time it identifies the clunkers. It turns out one third of the cars were bad - a huge number. But the other 2/3 were just fine and with the bad ones out of the mix, the market price for used cars rises from $4,000 to $9,000 on average (these aren't all the cheapest models so the average can be above $7,000). The government then sells the 66.67 million good cars at $9,000, raising $600 billion on the sale. So the government loses $100 billion overall.
But more importantly, after the trades are complete and people have their new cars, the economy revives, Bernson is a hero, and everyone lives happily ever after.
Or is that a fairly tale ending?