This analysis deserves to be fleshed out a little.
Timberland makes all its shoes in huge "shoe cities" in Vietnam and China. When those shoes are imported into the US they are counted in the trade balance, no?
Yet, when shoe sales go under, Timberland is going to take the hit just as much as the "shoe cites".
Which is to say, the GDP aggregate hides a lot of foreign activity, doesn't it?
I'm having a little trouble following the analysis for a couple of reasons. One is that trade impacts GDP so that $30B of imports lowers the balance and reduces it. But actually the equation in nat'l income stats is really:
GDP = (C-Cimp)+ (I-Iimp)+ (G-Gimp) + EX. Or GDP = C+I+G+EX-IM.
However the linkage may exist by another pathway since Savings - Investment = Net Exports (NX). In other words since we're running a huge and growing negative trade balance Savings is necessarily less than Investment.
As born out by the rising pools of foreign exchange reserves in China, et.al. that then get turned into US securities purchases. Wouldn't that pathway more likely lead to the correlation you see ?
Anyway a little food for thought. Thanks for putting this one on the table - a difficult and interesting conundrum as Uncle Allen would say. And very important.
You've established correlation, but I think that we have at least ATTEMPT possible explanation before we go to correlation. Is it simply cars and big ticket electronics? If that's so, there should be a better correlation with our trade deficit with Japan than our trade deficit with China. If that's not the case, I think we'd have to look more deeply at possible capitol flow reasons.
Well, the news isn't that good... it supports the notion that the US housing bubble may have been propping up the other property bubbles going on throughout the world. So a US housing collapse could become a global pheonomenon, and find it's way back to the US via exports and foregin holdings.
DaveL, I was trying to make it simple and just stick to the impact on GDP of MEW. With this simple approach, if there was no trade, any MEW related consumption (increase in C) would increase GDP.
No Trade:
GDP = C + I + G
Any increase in C increases GDP.
But with trade:
GDP = C + I + G + (Ex - Im)
If MEW related consumption flowed to imports, C would increase by the same as Im, and the changes in the two quantities would offset, so GDP would be unaffected.
I tried to stay from savings and investment!
Idaho-Spud and ac, I think this is the key question - and I think Roubini is correct that a housing led U.S. slowdown will impact the World (i.e. no decoupling).
CR - thanks, think I understand your line. Have to work a bit more on my getting it. Understand avoiding the S&I challenge but still think it's the mechanism - hence my agreement with you and Roubini that we can't be uncoupled.
Dear CR,
Alhough imports are subtracted from GDP accounting we should remember story of chinese made shoes. Expoted from Chına 5 Dollars FOB and bought by US customer at 60 Dollars. So 55 Dollars are still accounted in US GDP account.
I also claim many productivity gains in US are due to increased import penetration.
Let assume a final product consisting of 3 inputs and 1 assembly work each require 1 unit labor at a cost of 10 Dollars / unit.
In US final product is 40 Dollars and productivity is 10 Dollars for each work unit.
Now suppose we outsourced 3 inputs to China at a total cost of 10 Dollars and did final assembly in US.
Now final product cost is 20 Dollars and US labor productivity is 20 Dollars.
A productivity miracle of 100% !!!
I agree that there won't be decoupling if there is serious slowdown in US economy. But I still don't feel comfortable just believing so and I would like to prescience how will this slowdown transmit to the rest of the world. If MEW slows down, US imports will do so. A big part of the import bill is hold with China and Oil exporters that reinvest their surpluses in US securities. But also, the financial necessities of the US would not be so large. Thus, what would happen to the price of US issued securities? If long term rates creep up in the US, sure that their effects will be felt here in Europe. We will also feel the pain. I suspect that the coupling among long term debt markets is 100%. But I still don't see why US bond rates should rise very much.
May I make a guess as to your next post? As MEW slows, imports slow, our trading partners get fewer dollars, so they buy fewer MBS and Treasuries, raising the yield on both. Voila, higher interest rates.
BTW, I was surprised to learn that dollar volume of MBS exceeds Treasuries or corporate bonds. That is a huge market. Perhaps you could do a post on who is holding the MBS, and the fallout from foreclosures. I know that MBS is held by hedge funds, pension funds, Fannie Mae, our local banks. Do money market funds hold this too? We've all developed a safe feeling for money markets, but is that perhaps a misplaced perception of low risk? Couldn't money market funds lose 10-25% of value when MBS defaults rise?
Any MEW related consumption of imports does not impact GDP.
Well, if someone buys a big screen TV from Best Buy, made in China, and the markup is 50%, it would impact GDP! Also consider local shipping, taxes and installation and the impact could be substantial. The same would be true for a BMW.
Just eyeballing, but the trade deficit graph appears to show the opposite of the lag you suggested yesterday - it looks like the consumer spends first and extracts equity afterward.
As BS said, MEW has been important to foreign capital. Long term treasuries have been going lower for the last 3 weeks because they foretell a hard landing, which probably started in late August. I don't think the Foreigners bail if it is another 2001 type of landing, but they might if it turns out to be a 74-75 type landing.
If that happens, at some point in 2007-08 interest rates will rise to pre-Asian Flu levels, ending the artificially low rate era of 96-06. The short term pain will be bad, but if the US actually starts making some good decisions, maybe a long term benefit.
I suspect that consumers spend both before and after MEW. Spending before would take the form of debt consolidation, in other words paying off the credit card balances built up earlier. This might represent a collection fo relatively small ticket items, trips to Applebees, small consumer electronics, filling up the tank etc. I suspect spending after is more big ticket items, remodeling the kitchen, a new car, big screen TV etc. I'm not sure how you disentangle the two.
You seem to be suggesting a higher marginal propensity to import for cash raised through mortgage equity withdrawal than the average for all consumer spending. Is that necessary in order to get the chart you've shown us? MEW goes up, so spending goes up. Spending goes up, so imports go up. All done, without invoking a different marginal this-and-that. Wouldn't the correlation exist even without a higher marginal propensity to import for MEW money? It would make Mr Ocham happy if I'm right about this, and give us lots less 'splainin' to do.
I would also think that we can look to multiple causal mechanisms, with interest rates at the root of most or all. Lower interest rates lead to more MEW. Lower interest rates lead to higher home values. Higher home values lead to more MEW. Lower interest rates lead to more spending. MEW leads to more spending. Merely correlation between MEW and spending for some mechanisms, a causal link in others. That's the sort of complexity that leads to econometric food fights.
CR, how you tempt us with anticipation for your next entry......I don't want to spoil the ending but I betcha it has to do with debt securities
This analysis deserves to be fleshed out a little.
Timberland makes all its shoes in huge "shoe cities" in Vietnam and China. When those shoes are imported into the US they are counted in the trade balance, no?
Yet, when shoe sales go under, Timberland is going to take the hit just as much as the "shoe cites".
Which is to say, the GDP aggregate hides a lot of foreign activity, doesn't it?
Thanks for that CR. This does put another nail in the coffin for economic de-coupling of the US and foreign economies though, doesn't it?
I suspect Dr. Roubini is correct, that such de-coupling is not going to happen when the US consumer hits the ropes.
I'm having a little trouble following the analysis for a couple of reasons. One is that trade impacts GDP so that $30B of imports lowers the balance and reduces it. But actually the equation in nat'l income stats is really:
GDP = (C-Cimp)+ (I-Iimp)+ (G-Gimp) + EX. Or GDP = C+I+G+EX-IM.
However the linkage may exist by another pathway since Savings - Investment = Net Exports (NX). In other words since we're running a huge and growing negative trade balance Savings is necessarily less than Investment.
As born out by the rising pools of foreign exchange reserves in China, et.al. that then get turned into US securities purchases. Wouldn't that pathway more likely lead to the correlation you see ?
Anyway a little food for thought. Thanks for putting this one on the table - a difficult and interesting conundrum as Uncle Allen would say. And very important.
You've established correlation, but I think that we have at least ATTEMPT possible explanation before we go to correlation. Is it simply cars and big ticket electronics? If that's so, there should be a better correlation with our trade deficit with Japan than our trade deficit with China. If that's not the case, I think we'd have to look more deeply at possible capitol flow reasons.
Well, the news isn't that good... it supports the notion that the US housing bubble may have been propping up the other property bubbles going on throughout the world. So a US housing collapse could become a global pheonomenon, and find it's way back to the US via exports and foregin holdings.
DaveL, I was trying to make it simple and just stick to the impact on GDP of MEW. With this simple approach, if there was no trade, any MEW related consumption (increase in C) would increase GDP.
No Trade:
GDP = C + I + G
Any increase in C increases GDP.
But with trade:
GDP = C + I + G + (Ex - Im)
If MEW related consumption flowed to imports, C would increase by the same as Im, and the changes in the two quantities would offset, so GDP would be unaffected.
I tried to stay from savings and investment!
Idaho-Spud and ac, I think this is the key question - and I think Roubini is correct that a housing led U.S. slowdown will impact the World (i.e. no decoupling).
Best to all.
CR - thanks, think I understand your line. Have to work a bit more on my getting it. Understand avoiding the S&I challenge but still think it's the mechanism - hence my agreement with you and Roubini that we can't be uncoupled.
Dear CR,
Alhough imports are subtracted from GDP accounting we should remember story of chinese made shoes. Expoted from Chına 5 Dollars FOB and bought by US customer at 60 Dollars. So 55 Dollars are still accounted in US GDP account.
I also claim many productivity gains in US are due to increased import penetration.
Let assume a final product consisting of 3 inputs and 1 assembly work each require 1 unit labor at a cost of 10 Dollars / unit.
In US final product is 40 Dollars and productivity is 10 Dollars for each work unit.
Now suppose we outsourced 3 inputs to China at a total cost of 10 Dollars and did final assembly in US.
Now final product cost is 20 Dollars and US labor productivity is 20 Dollars.
A productivity miracle of 100% !!!
I agree that there won't be decoupling if there is serious slowdown in US economy. But I still don't feel comfortable just believing so and I would like to prescience how will this slowdown transmit to the rest of the world. If MEW slows down, US imports will do so. A big part of the import bill is hold with China and Oil exporters that reinvest their surpluses in US securities. But also, the financial necessities of the US would not be so large. Thus, what would happen to the price of US issued securities? If long term rates creep up in the US, sure that their effects will be felt here in Europe. We will also feel the pain. I suspect that the coupling among long term debt markets is 100%. But I still don't see why US bond rates should rise very much.
CR, another great analysis from you!
May I make a guess as to your next post? As MEW slows, imports slow, our trading partners get fewer dollars, so they buy fewer MBS and Treasuries, raising the yield on both. Voila, higher interest rates.
BTW, I was surprised to learn that dollar volume of MBS exceeds Treasuries or corporate bonds. That is a huge market. Perhaps you could do a post on who is holding the MBS, and the fallout from foreclosures. I know that MBS is held by hedge funds, pension funds, Fannie Mae, our local banks. Do money market funds hold this too? We've all developed a safe feeling for money markets, but is that perhaps a misplaced perception of low risk? Couldn't money market funds lose 10-25% of value when MBS defaults rise?
Any MEW related consumption of imports does not impact GDP.
Well, if someone buys a big screen TV from Best Buy, made in China, and the markup is 50%, it would impact GDP! Also consider local shipping, taxes and installation and the impact could be substantial. The same would be true for a BMW.
Just eyeballing, but the trade deficit graph appears to show the opposite of the lag you suggested yesterday - it looks like the consumer spends first and extracts equity afterward.
kaan and touche, Yes. But the portion that is imported isn't counted in GDP - but any domestic markup is included in GDP.
Schahrzad Berkland, good guess.
albrt, Its possible that the deficit leads MEW; the correlation is about the same with and without a lag.
Best to all.
As BS said, MEW has been important to foreign capital. Long term treasuries have been going lower for the last 3 weeks because they foretell a hard landing, which probably started in late August. I don't think the Foreigners bail if it is another 2001 type of landing, but they might if it turns out to be a 74-75 type landing.
If that happens, at some point in 2007-08 interest rates will rise to pre-Asian Flu levels, ending the artificially low rate era of 96-06. The short term pain will be bad, but if the US actually starts making some good decisions, maybe a long term benefit.
I suspect that consumers spend both before and after MEW. Spending before would take the form of debt consolidation, in other words paying off the credit card balances built up earlier. This might represent a collection fo relatively small ticket items, trips to Applebees, small consumer electronics, filling up the tank etc. I suspect spending after is more big ticket items, remodeling the kitchen, a new car, big screen TV etc. I'm not sure how you disentangle the two.
But aren't MEW and trade balance both connected to other factors like interest rates?
Dear Host,
You seem to be suggesting a higher marginal propensity to import for cash raised through mortgage equity withdrawal than the average for all consumer spending. Is that necessary in order to get the chart you've shown us? MEW goes up, so spending goes up. Spending goes up, so imports go up. All done, without invoking a different marginal this-and-that. Wouldn't the correlation exist even without a higher marginal propensity to import for MEW money? It would make Mr Ocham happy if I'm right about this, and give us lots less 'splainin' to do.
I would also think that we can look to multiple causal mechanisms, with interest rates at the root of most or all. Lower interest rates lead to more MEW. Lower interest rates lead to higher home values. Higher home values lead to more MEW. Lower interest rates lead to more spending. MEW leads to more spending. Merely correlation between MEW and spending for some mechanisms, a causal link in others. That's the sort of complexity that leads to econometric food fights.