July 31, 2009
Been down so long it looks like up
The Commerce Department reported today that the seasonally adjusted real value of the nation's production of goods and services fell at a 1% annual rate during the second quarter. That's about as bad as things ever got during the recession of 2001. But after the -5.4% and -6.4% growth rates that the Commerce Department now says characterized 2008:Q4 and 2009:Q1, some folks are cheering today's news. Reminds me a little of how I've seen people in Minnesota take off their shirts for the first 40oF day of spring, a little shocking to a traveler from San Diego.
On the other hand, our Econbrowser Recession Indicator Index is impressed with just how cold the winter was. This is a pattern recognition algorithm for identifying recessions that waits one quarter for new data and revisions before making a final assessment. Based on the 2009:Q2 GDP numbers just released, the value that the algorithm assigns to the first quarter of 2009 is 99.7, the highest reading since 1980. We'll declare the recession to be over when the index falls below 33.
In terms of specific factors contributing to the 2009:Q2 growth rate, consumption spending, housing, nonresidential fixed investment, inventory change, and exports each subtracted almost 1%-- had it not been for the positive contribution from falling imports and increasing government spending, the Q2 number would have been -4.3% instead of -1%. Should we be cheering the fact that falling imports were a key factor preventing GDP from declining even more? Falling U.S. imports can create problems for those countries trying to export to us and are a symptom of a very weak U.S. economy. But lower U.S. imports are a necessary element of our longer run adjustment process, and indeed, if the increase in U.S. private saving were just matched by the decrease in U.S. imports, we'd be exactly where we want to be in both the short and the long run.
Of course, we're not there at the moment, not even close, but I find some things to cheer about in the other components of GDP as well. It would take nothing more profound than for inventories to stop falling for GDP growth to be 1% higher over the next year than it was over the last. Indeed, it's reasonable to expect inventory rebuilding to make a positive contribution to GDP growth rates for the second half of the year. A similar calculation holds for housing. If the value of new homes built in the second half of the year remains at the same extremely low levels we saw in the first half, the GDP growth rate will be 1% higher. And it looks like home sales could well be higher, not just constant between 2009:H1 and H2.
On the other hand, it's hard to see nonresidential fixed investment making a positive contribution any time soon.
But by all means, go ahead, take your shirt off. As long as you understand that it's still plenty cold out there.
Posted by James Hamilton at July 31, 2009 10:56 AMdigg this | reddit
I have a question about GDP :
1) If an iPod is imported from China, and shipped into the US, it is counted twice, when Wal-Mart buys it, and when the end-user buys it. Correct?
2) If the same iPod drops in price from $200 to $100, with the features being the same, the contribution to GDP is halved, correct?
Thus, this leads me to conclude :
a) The Internet enables e-tailing, etc. which often cuts out one of the intermediaries in getting a product to an end-user. Thus, the disintermediation that the Internet creates between the manufacturer and buyer reduces GDP, even though it saves money for the customer.
b) Technological advancement only maintains GDP if the customer is spending the same dollars for a more advanced product. If that is flipped, where the customer is buying the same product for fewer dollars, GDP shrinks, even though the customer's purchasing power rose.
Technology can shrink GDP while simultaneously improving purchasing power and living standards. Thus, GDP per capita is starting to weaken as a measurement of prosperity, given the increasing diffusion of technology in the economy.
Posted by: GK at July 31, 2009 11:35 AM
I think this is the big news.
"It was just a horrendous result," said William O'Donnell, head of U.S. Treasury strategy at RBS Securities in Greenwich, Connecticut.
"It was the weakest bid-to-cover since September 2008, and by my numbers it was the biggest tail since February 1993. It was just a very, very weak result."
The tail indicates that dealers drove an unexpectedly hard bargain to raise yields, and lower prices, to buy the bonds. Ultimately, this could raise interest rates throughout the economy at a faster rate than might be appropriate given the lingering effects of the worst recession in decades.
"If rates unwind higher and too quickly � driven not by the Fed but by the old bond vigilantes � that will be the house of pain for all risk markets," said George Goncalves, head of fixed income rates strategy with Cantor Fitzgerald in New York.
Posted by: aaron at July 31, 2009 11:43 AM
GK -- No, the ipod is not counted twice.
GDP is based on final sales and does not include intermediate sales between wholesalers and retailers.
Posted by: spencer at July 31, 2009 12:37 PM
There is absolutely no doubt that the economy would have been better off today without the bailout and the omnibus spending bill. See my previous posts on the Damage of Government Spending and the True Cost of Government Spending.
Every dollar of tax revenues sucked out of the private economy and used as government spending costs the economy upwards of 50 cents. Every dollar of tax revenue kept in the private economy is either used as consumption which keeps the economy going in the short run (presuming they are consuming products and services produced by the private economy), or investment in capital and equipment and innovation, which is the sole source of productivity and growth in this economy - nay, in the world economy.
You need look no further than the overwhelming response to the "Cash for clunkers" program which was essentially a tax rebate on a durable good. Imagine where we would be today if the $787 billion "stimulus" program has been used as an across-the-board tax rebate? Better? Okay, now imagine how much better off we would be if the $787 billion had never been confiscated from the private economy in the first place! My point exactly.
I am not saying that the "Cash for Clunkers" was a smart program. On the contrary, it is a prime example of government manipulation our private lives in a way they think is politically correct. And it is a massive shell game fraught with deception and feel-good sound bites. If I own a clunker, that is my choice. If it pollutes and I choose to keep it and pay the costs of pollution, that is my choice. Clunkers may very well save the life of my 15 year old daughter, who just starting driving. I want her safe and alive should she have an accident. And that is MY choice, not the choice of OBAMA and gang. He can raise his children as he sees fit, and I will do the same, thank you very much. Now why don't the so-called economists who surround Obama cough up the data showing the REAL costs of this program to the American people: in addition to the price paid for the clunker, we also incur the costs of writing and monitoring the program, the car dealers' time and expense in implementing the program and filing the paperwork, and the collection and disposal (?????) of the clunkers themselves. And every one of those dollars spent began life as a dollar of OUR HARD-EARNED INCOME. My goodness, talk about your social engineering. What a joke.
Posted by: Sherry Jarrell at July 31, 2009 01:14 PM
OK, but the final conclusion still holds, merely to a lesser degree.
Also, the iPod would still be counted in both China's and the US's GDP, since it is made in China and shipped out, correct?
Posted by: GK at July 31, 2009 02:18 PM
You are absolutely right. Their political correctness is causing increasingly absurd distortions to economic policy. Not the least of which is 'global warming'.
On a totally separate note, I strongly recommend to any parent of a 15 y/o that you severely limit her access to a car until she is 18 (after which she will be off to college and doesn't need a car :) ). The ability to assess risk in a 16 y/o is MUCH less than in an 18 y/o, which in turn is much less than in a 20 y/o. The disproportionately high accident rates for drivers under 18 shows this.
If I had my way, the driving age would be 18, and the voting age would be 21. It is OK for the drinking age to be lowered to 18, in compensation.
Posted by: GK at July 31, 2009 02:24 PM
Imagine where we would be today if the $787 billion "stimulus" program has been used as an across-the-board tax rebate? Better? Okay, now imagine how much better off we would be if the $787 billion had never been confiscated from the private economy in the first place! My point exactly.
I hear you. For $787 billion, the government could have given a 6-month holiday on all Corporate and Personal Income taxes. This would immediately cause an economic surge (and the increase in sales taxes would have offset the cost partially anyway).
Posted by: GK at July 31, 2009 02:31 PM
Today's Q2 GDP numbers are a Rorschach inkblot that will tell you whatever you were already looking for in the direction of the economy. If you're looking for a Q3 turnaround, you like the slower rate of GDP decline. If you expect further GDP declines through 2009, you see it likely becasue of the drop in consumption and the increase in personal savings (implying more declines in consumption). Was it the Cheshire Cat who said you see it where you aren't looking?
Posted by: Mike Laird at July 31, 2009 03:01 PM
Mike - I do think that some are guilty of exactly what you are saying. But not me, not this time: I am not basing my opinion on today's numbers...no matter what the GDP growth rate today, it is my opinion, based on first principles and as much common sense as I can muster, that it would have been higher had the stimulus program never seen the light of day.
Posted by: Sherry Jarrell at July 31, 2009 03:11 PM
"Been Down So Long It Looks Like Up to Me" happens to be the title of a superb minor classic novel of the 1960's by Richard Farina (who died tragically just after the novel was published). Worth reading, particularly if you are a fan of Thomas Pynchon; who was a close friend of Farina - "Gravity's Rainbow" is dedicated to him.
Going beyond the immediate GDP numbers, a speculative question: Can a realistic model be constructed that posits an indefinite period of economic stagnation? That is an era when real GDP growth just about matches population growth resulting in little or no increase in per capita income?
I could think of various scenarios where production input and productivity grow slowly enough to cause that de facto stagnation - at least for the most mature national economies.
Just a thought...
Posted by: SecondLook at July 31, 2009 03:26 PM
"Imagine where we would be today if the $787 billion "stimulus" program has been used as an across-the-board tax rebate? Better? Okay, now imagine how much better off we would be if the $787 billion had never been confiscated from the private economy in the first place! My point exactly."
Hallucination? The $787B was/is/will be entirely borrowed to add to the fiscal deficit -- otherwise it is not stimulus.
Expectation matters. Without knowing that stimulus is on the way, (even though most not yet spent), the private economy may well have collapsed to a lower level instead of recovering.
Posted by: HZ at July 31, 2009 04:24 PM
Trough at May, considering trajectory of improvement, notwithstanding GK's otherwise consistent forecast (Lehman brought the trough forward, I think.)
Evidence suggests we're running up the back of the "V" (eg, actuals well in excess of all forecasts.)
Posted by: Steven Kopits at July 31, 2009 04:36 PM
GK: Yes, there are issues with the impact of technological and quality improvements in the measurement of GDP. Keep in mind however that as the prices of goods we purchase fall, our income goes further, and we can purchase more quantity, which helps to offset the impact on measured GDP. There is another significant issue with GDP that arises out of the fact that it is essentially an accounting measure of performance, a net income variable, which ignores any changes in the stock of wealth. Housing value is amortized; changes in stock market value are ignored. So it is an imperfect measure of U.S. wealth and performance, but these issues are well known. Alternative GDP measures have been floated and examined over the years but none do better on average. One final thought is that even if the annual measure is flawed, if it's equally flawed each period, at least the trend or change over time is meaningful and useful. So the growth rate of real GDP is fairly sound.
Posted by: Sherry Jarrell at July 31, 2009 08:54 PM
Just wondering: Have you adjusted your recession index to take into account the new methodology to generate GDP?
In an article elsewhere (can't remember the source), the change seems to have reduced GDP by some 13% except for the last few years when it reduced it by ~14-15%. I have no idea how the sectors, etc., are affected by the methodological changes.
You might want to back test your methodology on the new GDP data set to see if any changes are warranted. Not that I expect your 99 recession score to drop below a non-recession 33, but you may want to update and validate.
Sorry to create work for you. I do appreciate your efforts.
Posted by: Lilguy at August 1, 2009 07:10 AM
GK: As Spencer notes, GDP is based on final sales.
If the same ipod sells for a lower price, nominal GDP would fall, but real GDP would remain the same.
What really happens in your thought experiment is the price of ipods falls, but our nominal income doesn't fall by as much, because ipods are just part of the total economy. The end result is that we buy more ipods and more of everything else as a result of the cost savings. In general, improved sales technology that lowers the final price of that particular product would increase real GDP.
Posted by: JDH at August 1, 2009 07:12 AM
Lilguy: The recession indicator index graphed above was constructed with simulated real time data through 2004 (that is, the algorithm only used the data as it had been reported at that date to construct the number for that date) and "real" real time data since 2005 (that is, the number you see was publicly announced at the indicated date). That sample includes many different revisions much more dramatic than the most recent, and I believe the method is pretty robust with respect to such changes. Note that the index is calculated from GDP growth rates rather than levels.
Robustness with respect to data revisions is the entire objective here. Note we always wait for one full quarter's revisions before making the call, for example, the 2009:Q2 advance release is used to calculate the value of the index for 2009:Q1.
Posted by: JDH at August 1, 2009 07:24 AM
Sherry said: "There is absolutely no doubt that the economy would have been better off today without the bailout and the omnibus spending bill. ... Imagine where we would be today if the $787 billion "stimulus" program has been used as an across-the-board tax rebate? Better? Okay, now imagine how much better off we would be if the $787 billion had never been confiscated from the private economy in the first place!"
Given the fear caused by the magnitude of this recession and the massive job losses, I rather think that consumers would have saved this money or used it to pay down debts. I doubt that much of this $787 billion would have been spent by consumers. With the subsequent lack of demand, we might well be in the middle of Great Depression II by now.
Posted by: Alan at August 2, 2009 06:33 AM
GK: "...the iPod would still be counted in both China's and the US's GDP, since it is made in China and shipped out, correct?"
No - imports are subtracted from final sales when calculating GDP. So if it cost $50 to import from China and is sold for $200 in the US, $50 goes to Chinese GDP and $150 into US GDP. A lot of effort goes into making the figures as robust as possible.
You are right that continual changes in technology and price make it harder to make comparisons over time. As Sherry says, there is some attempt to reflect this both by attempting to take price changes out of the figures (comparing units shipped instead of dollars) and by using "hedonic" measures which reflect the increase in quality of goods and offset that against price rises. It's not a perfect science though.
Posted by: Leigh Caldwell at August 2, 2009 07:54 AM
Been down so long it looks like up is really a good way to put the current responds from governments around the world.
Don't worry, the numbers are going up or at least slowing down, is what you hear these days often. Everything will be better by 2010 or by End of 2009.
The irony, if there is any, is the numbers are not even close to anything before the meltdown. Millions will be without jobs, income and home. It will take a much longer time to get all these people back into a job.
If now these billions or trillions of Dollars spend on stimulus packets are well spend, without corruption, is totally unknown. Also how this money ever will be recovered.
I see no alternative to this stimulus programs. I do not see it as bad, BUT my concern is who will control how this money is spend. Will it really reach the programs or will it lend in the pockets of certain interest groups and individuals?
It can all work without to many casualties, I want to believe, but I have second thoughts.
Consumers for now will find it difficult to spend money. Banks are not willing to lend money to consumers or businesses with ease, even so the same banks have received billions of taxpayers money to stay in business.
Posted by: August Ashley at August 2, 2009 09:13 AM
Actually, GDP is not based on final sales, but on domestic value added. Meaning that an IPod manufactured in China and sold in America is counted twice, with the price Wal-Mart sells it for being a positive contribution and the price Wal-Mart buys it for being the negative contribution. Wal-Mart's margin, its value added, is the net contribution to GDP.
Also, since IPod is owned by Apple, the profits from its Chinese operations will add to American national income, though that could be classified as factor income from abroad rather than GDP.
Posted by: Stefan Karlsson at August 2, 2009 09:29 AM
Sherry said: "There is absolutely no doubt that the economy would have been better off today without the bailout and the omnibus spending bill..."
What you should have said is "There is absolutely no doubt IN MY MIND..." Because trust me, there are plenty of people out there who don't subscribe to the economic school of Norquist. It surprise you to learn that we even have reasons for holding these opinions.
Posted by: John at August 2, 2009 12:05 PM
One thing that struck me that no one is mentioning isa the drop in the deflator from 1.9 o 0.2% q/q. Thoughts on this?
Posted by: s at August 2, 2009 01:17 PM
Alan, paying down debt may not contribute to GDP directly, but it allows for future growth, by decreasing risk and allowing for more money to be spent and invested in the future. The result is not immediate, but a higher overall growth rate is much more valuable than a one time bump.
Posted by: aaron at August 2, 2009 08:28 PM
The stimulus did effect the overall GDP rate by about 3% as even professor attested last week.
Here's a way to think about what the impact of ARRA has been on 2009Q2 growth. About $60 billion of stimulus funds had been expended by end-June, of which a large portion is in the form of tax rebates. The price deflator is about 10% higher in 2009Q2 than in 2005, so $60 billion translates to about 54.5 billion 2005$. This is a cumulative figure, while 2009Q2 SAAR GDP was 12892 billion, or 3223 billion Ch.2005$ at quarterly rate. If the multiplier is 0.5 (keeping in mind a large chunk of these funds are tax rebates), then growth was about 3 percentage points higher (q/q SAAR) than would have otherwise occurred; I think this is how Josh Bivens arrived at the conclusion that GDP growth would have been 3% lower in the absence of the ARRA.
I know this is not what you want to hear Sherry, but those are the numbers derived from a number of Phd economists.
When you say that there should have been a 6 month tax holiday for a better use of the money, you are flat out wrong. The MPC numbers bear out, and Prof Chinn went through these numbers last October.
The $ change in GDP due to tax cuts is less than 1, and corporate tax cuts are far less than $1 or around $0.30. A tax holiday is the best of the bunch, at 1.29, which is partially what you proposed, but when combined with corporate tax cuts, you most likely are around $1. So in essence, you get more out of the stimulus in terms of GDP growth with a stimulus package than a tax cut provided it gets out there.
Furthermore, not only are you increasing the GDP by more, but you are creating more jobs because of a larger GDP growth. You do have faith in Okun don't you?
When you cut taxes, you may want to argue that long term growth is the goal, but you can not argue that short term growth is better off, because it simply isn't. The theory not only proves you incorrect, the numbers coming in right now do, which was evident by our 1% contraction which beat estimates.
This is from an engineer mind you, and not an economist. If I can figure out this stuff, someone with your knowledge surely should be able to. Don't let politics cloud your opinion in science or economics. It is based on numbers and evidence, not the beliefs of a party. look at the data, then make a supposition, don't fit the data around your supposition. This is something that goes across all fields.
Posted by: Mike at August 14, 2009 11:33 AM