A modestly brighter GDP report

The Bureau of Economic Analysis reported today that U.S. real GDP grew at an annual rate of 3.2% during the fourth quarter of 2010. That’s about the historically average growth rate. But we expect much better than average at this point in the cycle, and need much better than average to make real progress with the unemployment rate.

The latest GDP numbers bring our Econbrowser Recession Indicator Index for 2010:Q3 down to 5.3%. There’s no question that we’re in the expansion phase of the business cycle, and have been for some time, though it’s only with this new report that the level of real GDP is back to making a new all-time high.



The plotted value for each date is based solely on information as it would have been publicly available and reported as of one quarter after the indicated date, with 2010:Q3 the last date shown on the graph. Shaded regions represent dates of NBER recessions, which were not used in any way in constructing the index, and which were sometimes not reported until two years after the date.
rec_ind_jan_11.gif



Consumption spending was strong in the fourth quarter, and could have generated essentially all the real GDP growth by itself. Exports added another 1%. Since imports are subtracted from GDP, the fourth-quarter decline in imports would have provided a further 2.4% boost to the reported GDP growth rate. Nonresidential fixed investment contributed 0.4%, almost entirely from equipment and software. And even housing made a slightly positive contribution.




gdp_comps_jan_11.gif


With all these strong positives, how did we end up with only 3.2% growth? The answer is that a huge estimated decline in inventories subtracted back out 3.7%. The extra spending by consumers and firms was much more than we produced domestically, and the difference represents goods sold out of inventory.

But the fact that a huge negative contribution of inventories coincided with a huge positive contribution of imports does not seem to be a coincidence. There’s a clear pattern in the recent data that when one of these makes a positive contribution to GDP growth, the other makes an offsetting negative contribution. Although we often think of inventories as a substitute for production (you could either produce a good or sell it out of inventories), in the current environment inventories seem to act more as a substitute for imports (you could either import the good, or sell it out of inventories). So although inventories shouldn’t be the same drag on GDP in 2011, I expect imports to go back up and exert a drag of their own.



Contribution to annualized real GDP growth of change in imports (black line) and change in inventories (red line), 2001:Q1-2010:Q4.
imp_inv_jan_11.gif



My bottom line: the recovery continues to gain momentum, but we still have a long, long way to go.

StumbleUponLinkedInReddit

29 thoughts on “A modestly brighter GDP report

  1. flow5

    No, money works quickly. Where are the nominal gDp target proponents? Just pissing in the wind?
    Real-output is set to outstrip the supply of goods & services offered in the market place (i.e., at high enough run rates to generate new, higher price levels).

  2. Anonymous

    JDH wrote:
    But the fact that a huge negative contribution of inventories coincided with a huge positive contribution of imports does not seem to be a coincidence. There’s a clear pattern in the recent data that when one of these makes a positive contribution to GDP growth, the other makes an offsetting negative contribution. Although we often think of inventories as a substitute for production (you could either produce a good or sell it out of inventories), in the current environment inventories seem to act more as a substitute for imports (you could either import the good, or sell it out of inventories). So although inventories shouldn’t be the same drag on GDP in 2011, I expect imports to go back up and exert a drag of their own.
    A very good point professor. It does point to another weakness of the GDP as an indicator and why it is so often misinterpreted. There is much in the GDP that is not a precise indicator of real economic activitiy. You have a similar trade off between the private (productive) and public (confiscatory) components of the GDP.

  3. Ivars

    More like the USA coming closer to be in line for popular revolts. With the Middle East under revolutions as first political outburst of the Great Recession; and consequent permanent rising trend in Oil prices, how will American Economy fare in 2011-2012? Double dip is imminent in end of 2011-Q1 2012.
    1930:March 2 – Mahatma Gandhi informs the British viceroy of India that civil disobedience will begin 9 days later.

  4. rootless

    “My bottom line: the recovery continues to gain momentum, but we still have a long, long way to go.”
    Sorry, but I doubt the validity of the first half of the statement from a scientific point of view.
    Considering that, according to BEA, the magnitude of the revision from the preliminary release of the real GDP change to the latest release amounts to 1.3 %, and the standard deviation of the revisions amounts to 1 %, I don’t see how any valid conclusion about a gaining momentum of the economy could be drawn from these preliminary GDP data. Using the standard deviation from the first to the latest revision, the one sigma confidence interval around the current release ranges from 2.2 to 4.2%. The two-sigma confidence interval (about 95% probability of statistical significance) ranges from 1.2 to 5.2%. With this uncertainty in the preliminary GDP change data, the real GDP change is statistically not yet distinguishable from the previous quarter data. After all the revision have been done, the final outcome can be everything, from a strong acceleration to a deceleration of the economic growth compared to the last quarter.
    Also, it appears that the largest magnitude of the revisions is to be found after the “final” GDP release on average, starting with the first annual revision. So we won’t be able to say anything about the current direction of the economy without a high degree of uncertainty before summer, will we?
    The inverse correlation between import change and change in inventory change is interesting. Why would inventory change fluctuations mostly occur for imported goods, but not for domestically produced ones? I’m puzzled.

  5. colonelmoore

    Can someone explain why inventory levels are correlated with imports and not production? Or is this something that has been observed but is not understood?

  6. Anonymous

    @Anonymous at January 28, 2011 10:17 AM:
    You wrote:
    “A very good point professor. It does point to another weakness of the GDP as an indicator and why it is so often misinterpreted. There is much in the GDP that is not a precise indicator of real economic activitiy.”
    I don’t see what you apparently see. Why do the subtraction of imported goods and the addition of the inventory change to obtain the GDP (or the apparent inverse correlation of their changes?) point to that GDP was not a precise indicator of real economic activity? What is wrong with these two variables in the GDP formula?

  7. aaron

    Colonolmoore,
    Domestic production for domestic consumption is likely to be more JIT. I’m not sure we produce a lot of physical consumer products other than food.
    Of couse, with cheap real estate and labor that may change. But knowing that most people don’t have money to buy things, neither seem cheap.

  8. Brian

    Oh dear, Mr. Hamilton, I’m afraid you’re getting way ahead of yourself here. A recovery? Considering the booms we had in the 90’s and 2000’s, you’re actually going to call this a recovery? Good luck with that thesis. And there are many out there that share your view, and many business owners that I know are expecting a great economy in 2011, but unfortantely, just on the track record of the consensus view, 2011 will be anything but good.

    I’ll continue to sit here on the sidelines with a pile of cash waiting for an entry to short the market–have you seen these crazy valuations? In the face of cpllapsing profit margins? And the banks look on pace for another crisis late this year, early next year.

    How do you propose a real recovery with rising resource costs and high unemployment? That’s a squeeze on profit margins, and bottom line: profit margins are the only things that matters when it comes to business hiring.

    And let’s look a what corporations are doing: hoarding cash? No business wants to hoard cash–and unlike how it’s being spun, corporations hoarding cash is NOT a good thing. What they’re basically indicating is that they see no way of using that money to grow their business (and they’re hoarding cash because they are expecting another banking crisis, and back in ’08 they were having trouble in the money market accounts–a mistake they don’t plan on repeating).

    Well, Mr. Hamilton, I hope you’re correct in proposing that the “recovery” will continue, or that we’re even in a “recovery”–but for crying out loud: I bought the VIX for under $17!

  9. 2slugbaits

    JDH: There’s a clear pattern in the recent data that when one of these makes a positive contribution to GDP growth, the other makes an offsetting negative contribution.
    So does this mean we should expect you or Menzie will treat us with another error correction model in the next few days? ;->

  10. rootless

    There is another question that has come to my mind. If the change in private inventory change has such a large import component, doesn’t this mean that the BEA’s number for the change in “real final sales of domestic product” (+7.1% in Q4 2010) was flawed to a large magnitude?
    According to BEA, “real final sales of domestic product” is calculated by subtracting private inventory change from GDP. However, as I see it, only the domestically produced component of the inventory change shall be subtracted from the GDP, if one wants to calculate the final sales of domestic product correctly. Otherwise by subtracting the total private inventory change, its import component is subtracted twice, since it already is part of the import component in the trade balance part of the GDP formula.
    What don’t I see here?
    BTW: I don’t think one can conclude from the inverse correlation between the percentage contributions to GDP of the import change and the change in inventory change that there is a substitution going on between the both. This also could be due to just a similar, but inverse behavior of the two variables within the business cycle, without a direct causal link between the two. And the similar magnitude of the changes could be just a spurious result of the current proportions of imports and change in private inventories as part of the GDP.

  11. Mark A. Sadowski

    Final sales of domestic product FSDP) is a preferred measure of aggregate demand by some Quasi-Monetarists (nominal GDP and final sales to domestic purchasers are two others). FSDP rose an astonishing 7.3% at an annual rate in the fourth quarter (since the recovery began it has never exceeded 3.0%). In real terms the increase was 7.1%, the most in 26 years. I think this shows a number of things:
    1) QE2 worked
    2) Expectations matter (Bernanke’s Jackson Hole speech probably did more than the actual implementation)
    3) There are no lags to monetary policy.
    Needless to say an explicit target would help much more than an arbitrary asset purchase figure. But first more people need to be convinced that 1) printing money has real effects in the short run, and that 2) something needs to be done.

  12. aaron

    Mark, two things.
    Seasonalized?
    I suspect this winter and holiday season people, like myself, are feeling thrift burn-out. No income improvement, just reduces savings that’ll need to be made up for later.
    Make that three. How many people stopped paying mortgages or rent providing a temporary boost in cash?
    Err… Four. Banks have already written off a lot of mortgages, so principal is profit right now.

  13. Mark A. Sadowski

    Aaron,
    Seasonalized? The data is all seasonalized so I completely fail to comprehend your point.
    In any case ask the same questions in three months. (Do you really think producers will not replenish greatly depleted inventories at some point?) Eventually you’ll have to concede that it worked.
    P.S. If people are no longer paying mortgages or rent what are they doing for shelter this bitterly cold winter? Just wonderin’.

  14. aaron

    There was no point, just didn’t bother to check myself.
    I’ll ask in 8 months.
    In the mean time, I’ll short gold.
    Everyone will get excited, but without dealing with the mortgage bubble and an energy policy, we won’t have a real recovery.

  15. Mark A. Sadowski

    Aaron,
    And I suppose by you reasoning renting or buying another place frees up money to spend on other things, but exactly how?
    Do I detect a lack of rationality?

  16. rootless

    @Mark A. Sadowski:
    “FSDP rose an astonishing 7.3% at an annual rate in the fourth quarter (since the recovery began it has never exceeded 3.0%). In real terms the increase was 7.1%, the most in 26 years.”
    How do you even know this number is what it is said it represents? If the BEA calculate Final Sales of Domestic Product how they say they calculate it, the 7.1% could be greatly exaggerated, since then they haven’t subtracted the import component of the private inventory change from the total private inventory change before subtracting the remainder, i.e., the domestic component, from the GDP to obtain the final sales. If the import component of the change in private inventory change amounts to a significant fraction of the total change and if imports decrease strongly, like they have in recent quarter, then the import decrease artificially inflates the change in the final sales number.
    Or, please could anyone finally explain to me what I don’t get here and where my thinking goes wrong?
    “I think this shows a number of things:
    1) QE2 worked
    2) Expectations matter (Bernanke’s Jackson Hole speech probably did more than the actual implementation)
    3) There are no lags to monetary policy.”
    No, it doesn’t show either of these. Beside the objection above, correlation doesn’t prove causality.
    What is the causal chain through which the expansion of the monetary base by a hundred billion dollars or so from the start of QE II in November 2010 to the end of the year has supposedly increased sales of domestic product in Q4 2010? Why would the addition of another hundred billion dollars to the bank reserves suddenly have an effect, which the more than 1 trillion of reserves that already had been accumulated and idling before hadn’t had? Plausibility of the alleged effect is missing.

  17. Mark A. Sadowski

    aaron,
    These days shorting gold might be a good bet.
    However, how does one buy another place before defaulting on the previous one? Or is that too impolite to ask?

  18. Mark A. Sadowski

    rootless,
    You wrote:
    “Plausibility of the alleged effect is missing.”
    The effect exists. Your inability to explain it away in a plausible manner also exists.
    Being an Austrian must really suck these days.

  19. aaron

    No. Not at all. Sorry if I seem argumentative, that wasn’t my intent (it rarely is). I’m just not very good at putting my thoughts to words. I tend to be brief to keep my focus. And the new Andoid text input system really, really sucks.
    Moving to a less expensive apartment frees up income. Moving to a less expensive home frees up income.
    Strategic default was not uncommon after prices started to fall. Probably doesn’t happen this way much now, but early on some people would buy a house they intend to stay in and then default on their previously purchased, overpriced house and other debts (the occupied home being protected in bankruptcy). It would require having cash or credit availible for another house and income to make the monthy payments on both and faith that they won’t need credit for quite a while.

  20. aaron

    In many places a person can rent or buy a nicer property with a lower payment than their bubble mortgage. They just need to be willing and able to walk away.

  21. flow5

    As I can’t dissect the final product data, I watch the other side of the equation. There’s no such think as “long & variable” money lags. Real-output will jump in the 1st & 2nd qtrs. Because the deflator won’t follow along, there’s no need to restrict nominal gDp. I.e., the lesson is to target the price level.

  22. Jeff

    I don’t know how the NIPA numbers are actually put together, but suppose for a minute that

    1. inventories are not actually measured, but calculated as the cumulative sum of (income less final sales), and
    2. imports are measured with an error, and
    3. everything else is measured perfectly.

    Then since net exports are part of income, a positive measurement error in imports induces an underestimate of income, but since final sales are measured perfectly, this implies that the change in inventories has the same negative error.

    The result is a negative correlation between the change in inventories and imports. It has no economic meaning, it’s just an artifact of how the numbers are calculated.

  23. Mark A. Sadowski

    rootless,
    Someone who’s opinion I value pointed to your comment that the final sales of domestic goods number may be flawed. In the spirit of the pursuit of truth lat me offer the following.
    Preliminary statistical investigation reveals that you are right to be skeptical. The correlation between the imported goods and inventory contribution to GDP is reasonably high and I estimate the elasticity to be about -1. If so a more accurate estimate of FSDP result would be about 5%.
    So, yes the BEA FSDP figure appears to be highly flawed. (Sigh.)

  24. rootless

    @Mark A. Sadowski:
    You replied to me:
    “You wrote:
    ‘Plausibility of the alleged effect is missing.’
    The effect exists. Your inability to explain it away in a plausible manner also exists.”
    I think I have provided an argument, why the effect claimed by you wasn’t plausible. More than 1 trillion vs. about 100 billion. And I don’t have the burden of disproving your claims. It’s you who has the burden of proof for your claim that the effect exists, or the burden of explaining the causal chain through which QE II has suddenly increased final sales as much, supposedly, despite the fact that the previous more than 1 trillion US-dollars added to the bank reserves hadn’t any such effect before. Resorting to solely ad hominem doesn’t do it.
    “Being an Austrian must really suck these days.”
    That may be, but I don’t care. What makes you think I was an “Austrian”?

  25. Dave Cohen

    From John Mauldin’s A Bubble In Complacency, as quoted in today’s DOTE post The Propaganda Blitzkrieg Begins

    You remember our old friendly equation:

    GDP = Consumption + Investments + Government Expenditures + (Net Exports)

    Now to get Real GDP (actual GDP after inflation) you have to take away the effects of inflation / deflation. This is done by the use of a deflator built in for each category. But the deflator for exports / imports is a little tricky at times.

    Moody’s correctly noted that “private inventories were an enormous drag on growth” and concluded that this was a good thing, in that they assumed that meant inventories went down and thus inventory rebuilding in future quarters will add to GDP growth. And that is where you have to look at the numbers, and there we find our anomaly. There really wasn’t that big a drop in inventories. It was in large part in the statistics, not in the warehouse.

    Oil in the 4th quarter rose from roughly $81 to $89, or about 10%. On an annualized basis, this is 40%. Inventory investment is equal to the change in book value of the inventories, minus what is known as the IVA, or inventory valuation adjustment, which is used to correct for prices going up or down. Because the value of oil rose and thus cost more to acquire, the accounting requires that you reduce the value of the current inventories. Thus “real” imports fell at a 13% annual rate. Why? Because the deflator rose by 19%, largely because of the rise in the price of oil.

    I know, I know, I just wrote that because the price of oil went up, the “real” value of imports went down, as well as inventories. Some of you are getting economic whiplash right about now.

    If oil were to go back down this quarter by the same amount, that “growth” could be wiped out. There is no conspiracy here. It is just a statistical necessity… How much did it change things? Lacy Hunt thinks by anywhere from 0.5% to 1%. That means GDP is still a positive number, but there is not a “3” handle at the beginning of it. In the grand scheme of things, no big deal, as it will balance out over the coming quarters and years. But I just wanted to point out (once again) that you have to take some of the numbers we get from our government with a few grains of salt.

Comments are closed.