Claim: Cash-For-Clunkers MASSIVELY Distorted GDP

[Puppycow]

I want to note the following article here as it makes some interesting assertions as well as some testable claims.

CHART OF THE DAY: Cash-For-Clunkers MASSIVELY Distorted GDP

If anyone mentions the just-released 3.5% U.S. third quarter GDP growth, just throw this chart in their face. Cash for Clunkers clearly distorted the U.S. economic figures in an unsustainable fashion.

According to the Bureau of Economic Analysis (BEA), motor vehicle output spiked a seasonally-adjusted 157.6% quarter on quarter. This is completely unprecedented. Vehicle output is clearly going off a cliff next quarter. The question will be how low can the blue line below go.

(Can't quote much more as it is a rather short article altogether.)

According to another article:

Motor vehicle production contributed 1.7 percentage points to growth, nearly half of the total GDP increase.

So the basic facts seem to be correct.

The basic empirical question is did cash-for-clunkers "prime the pump" so-to-speak, or simply rob sales from the future?

BTW, how much is 1.7 percentage points of the quarterly GDP? The second article gives an annualized figure of $14.3 trillion for GDP. 1/4 of that is $3.575 trillion and 1.7% of that is $60.74 billion. Remember that the government spent $3 billion on cash-for-clunkers. If a $3 billion program can really MASSIVELY "distort" the GDP, then maybe it was actually rather effective?

The testable claims are:
1) "Vehicle output is clearly going off a cliff next quarter."
2) "Q4 is likely to be very weak"
3) (Combination of 1 and 2) "Next quarter, not only are we unlikely to get Q3's boost, but motor vehicle output data could subtract from GDP as well. So watch out for the cliff..."

 

[Just thinking]

Well, just ask yourself ... where's the "growth" ?

If all or most of what Cash-for-Clunkers did was sell off overstocked inventories, then it's not at all unreasonable to assume that there really was little to no growth in that sector. Major automobile manufacturers had many months of very poor sales numbers ... some even went bankrupt. (I can easily imagine lots and lots of new unsold cars just waiting.) Have they ramped up production by 3.5% ? Are auto parts suppliers increasing their output to match the increased sales from the program? Any new manufacturing plants being considered ... or more plant closings? Has there been an increase in workers hired to justify any real growth? Also, just how analogous is this to first-time home owner's incentives? How much did that also contribute to the GDP number? Has new home production increased? ... last I heard, it was down. How many of these sales were on foreclosures, which are quite abundant?

 

[YoPopa]

BTW, how much is 1.7 percentage points of the quarterly GDP? The second article gives an annualized figure of $14.3 trillion for GDP. 1/4 of that is $3.575 trillion and 1.7% of that is $60.74 billion. Remember that the government spent $3 billion on cash-for-clunkers. If a $3 billion program can really MASSIVELY "distort" the GDP, then maybe it was actually rather effective?

Correct me if I am wrong, I'm no mathematician. I think you are calculating 1.7% of the total GDP for the quarter. You should be figuring that 1.7% based on the growth of the GDP which would be a much smaller number. If the program were able to MASSIVELY "distort" the GDP to the extent you calculate then I fear the effect of withdrawal would be even worse.

 

[Puppycow]

Correct me if I am wrong, I'm no mathematician. I think you are calculating 1.7% of the total GDP for the quarter. You should be figuring that 1.7% based on the growth of the GDP which would be a much smaller number.

I don't think so. Here's the relevant quote again:

Most of the clunker sales came out of inventories, but production of vehicles rebounded smartly after a sharp pullback earlier in the year. Motor vehicle production contributed 1.7 percentage points to growth, nearly half of the total GDP increase.

If the program were able to MASSIVELY "distort" the GDP to the extent you calculate then I fear the effect of withdrawal would be even worse.

That's basically the claim of the article. I'm doubtful. I'm sure there will be a drop off, but not that we will "fall of a cliff" so to speak.

 

[Francesca R]

Motor vehicle production contributed 1.7 percentage points to growth, nearly half of the total GDP increase.

If all or most of what Cash-for-Clunkers did was sell off overstocked inventories, then it's not at all unreasonable to assume that there really was little to no growth in that sector.

If domestic vehicle output contributed 1.7% to growth, which seems to be the case, then that is separate from inventory rundowns or imports. However, some of it would probably have happened anyway, since it had collapsed in prior quarters

BTW, how much is 1.7 percentage points of the quarterly GDP? The second article gives an annualized figure of $14.3 trillion for GDP. 1/4 of that is $3.575 trillion and 1.7% of that is $60.74 billion.

GDP maths is fiddly but that's approximately correct. So that is the increase in real spending on vehicles between July and September, some of which was probably caused by stimulus. From what I read, most of the rest of the GDP increase was due to policy stimulus announced in the Feb 2009 bill/act (which cash for clunkers wasn't part of).

That's the way fiscal support is supposed to work--the government spending money in a way that incentivises consumers to spend more, and now. As such it usually unwinds itself later. If it is borrowing from the peak of the next upswing, that's not necessarily a bad thing at all, particularly if it helps provide the next upswing in the first place. If there's never going to be a next upswing, then it's arguably providing growth today at the price of contraction later.

The stimulus could--of course--have been oriented towards investment (rather than boosting consumption) with a longer term slow-burning payoff. Then commentators would probably have been writing about another quarter of recession and how bad that was.

 

[Aepervius]

Around here in germany, after the same type of operation than cash for clunker, the auto industry & sales plunged. Maybe somebody can dig the data of this trimester , my google fu did bring only old data, but that should be an indication of what could happen in the US soon.

 

[Just thinking]

If domestic vehicle output contributed 1.7% to growth, which seems to be the case, then that is separate from inventory rundowns or imports. However, some of it would probably have happened anyway, since it had collapsed in prior quarters.

I am curious as to just what to make from this paragraph in your linked article ...

Most of the clunker sales came out of inventories, but production of vehicles rebounded smartly after a sharp pullback earlier in the year. Motor vehicle production contributed 1.7 percentage points to growth, nearly half of the total GDP increase.

If most of the auto sales were from existing inventories (which I initially thought was the case), and the program helped the GDP by giving 50% to its value (1.7%), then how can production also be a gain of 1.7% to the growth? It doesn't seem to add up.

 

[drkitten]

Most of the clunker sales came out of inventories, but production of vehicles rebounded smartly after a sharp pullback earlier in the year. Motor vehicle production contributed 1.7 percentage points to growth, nearly half of the total GDP increase.

If most of the auto sales were from existing inventories (which I initially thought was the case), and the program helped the GDP by giving 50% to its value (1.7%), then how can production also be a gain of 1.7% to the growth? It doesn't seem to add up.

Because the sales depleted existing inventories, which had to be replenished via production.

If I sell out my entire stock of widgets, that comes from existing inventories. But unless I then shutter my widget store and go retire to a sunny beach somewhere, I need to buy more widgets from the factory.

And one of the advantages of selling out my widget line is that I can re-stock with something better suited to the current widget sales climate.

 

[Francesca R]

I am curious as to just what to make from this paragraph in your linked article ...

I didn't link one, but the GDP report in detail is here

If most of the auto sales were from existing inventories (which I initially thought was the case), and the program helped the GDP by giving 50% to its value (1.7%), then how can production also be a gain of 1.7% to the growth? It doesn't seem to add up.

This is confuddled (like I said, GDP maths is fiddly). When they say motor vehicle output made a contribution of 1.7% to GDP growth, it means that the growth in production of vehicles alone accounted for 1.7% (annualised) growth in total economic output, which is something like $220 billion (annualised), or $55 billion in the three months. You divide it by the base level of GDP (just under $13,000 billion) to get the 1.7%. Motor vehicle output itself grew much more than 1.7% from what its own base was, since vehicle production is a small fraction of total economic production.

Then, final sales of vehicles is not the same as output, since sales can come from inventories as well as production, and from imports. And it is the change in sales that matters, not how big they are. And where it gets even more fiddly is that with inventories, it is the change in the change in inventries that determines their contribution to growth. (If you run down inventories, but by less than you ran them down last quarter, then the contribution of inventory accumulation/depletion to GDP growth is positive.) This might help

 

[Just thinking]

Because the sales depleted existing inventories, which had to be replenished via production.

If I sell out my entire stock of widgets, that comes from existing inventories. But unless I then shutter my widget store and go retire to a sunny beach somewhere, I need to buy more widgets from the factory.

And one of the advantages of selling out my widget line is that I can re-stock with something better suited to the current widget sales climate.

Right ... but is that growth, or just status quo? I'm not arguing that it's not good to sell off the backed-up inventory and then replace it ... I'm curious where the growth is. From the numbers it looks like 1.7% of GDP out (sales) matched with 1.7% in (production).

 

[Francesca R]

The growth is that the US produced more vehicles than the previous period, sold more, and ran down inventories by less.

 

[Just thinking]

(If you run down inventories, but by less than you ran them down last quarter, then the contribution of inventory accumulation/depletion to GDP growth is positive.)

So basically I can still make more cars than I can sell (goes into inventory), but if it's less going into inventory than last quarter ... I'm growing GDP?

 

[Just thinking]

The growth is that the US produced more vehicles than the previous period, sold more, and ran down inventories by less.

You realize this can still leave one in negative territory ... right? In other words, still losing money (depending on where those cars that were sold came from).

 

[kevinquinnyo]

Don't forget the government destroyed several billion dollars worth of vehicles. They literally destroyed them.

And we paid for them to do this.

How about we destroy old apartment complexes too, and subsidize developers to build new ones?

Why don't we just destroy say, the entire city of Detroit and subsidize new construction?

It will provide a boost to the construction industry.

Or I've got a better idea, why don't we start infecting people with diseases to boost the medical industry?

...

 

[Malerin]

How about we destroy old apartment complexes too, and subsidize developers to build new ones?

http://en.wikipedia.org/wiki/Pruitt-Igoe

http://en.wikipedia.org/wiki/Cabrini-Green

http://atlantahousingauthority.blogspot.com/2008/07/public-housing-projects-in-atlanta-are.html

 

[GreyICE]

It's frequently the most cost-effective solution. There's some skyscrapers in Manhattan that are so inefficient, people would be better off tearing them down and starting over (10 year payback period).

 

[psychictv]

Don't forget the government destroyed several billion dollars worth of vehicles. They literally destroyed them.

Technically just the engines were destroyed.

 

[Dr Adequate]

If anyone mentions the just-released 3.5% U.S. third quarter GDP growth, just throw this chart in their face. Cash for Clunkers clearly distorted the U.S. economic figures in an unsustainable fashion.

So ... the stimulus package stimulated the economy?

Damn, can't Obama get anything right?

 

[psychictv]

So ... the stimulus package stimulated the economy?

No, no, don't you see? It distorted the economy.

 

[kevinquinnyo]

http://en.wikipedia.org/wiki/Pruitt-Igoe

http://en.wikipedia.org/wiki/Cabrini-Green

http://atlantahousingauthority.blogspot.com/2008/07/public-housing-projects-in-atlanta-are.html

I know for a fact that the federal government has actually done this before, and of course, it failed spectacularly, but these are just examples of federal housing projects failing spectacularly, and then failing again.

My point is that destruction of wealth is never a good idea.

Imagine if Cash For Clunkers had been done privately by say the Alliance of Automobile Manufacturers. Do you think there would be any reason whatsoever to destroy perfectly good vehicles?

Of course not. They would sell them.

Destruction is never beneficial. Only the government has the ability to do something so negligent.