Brad Setser

Follow the Money

Cross border flows, with a bit of macroeconomics

Print Print Cite Cite
Style: MLA APA Chicago Close


This seems quite bad …

by Brad Setser
January 2, 2009

There is a real risk that the worst financial crisis since the 30s will lead to the sharpest global downturn since the 1930s. The latest ISM survey is rather grim. The New York Times reports:

“Manufacturing activity continued to decline at a rapid rate during the month of December,” said Norbert J. Ore, chairman of the Institute for Supply Management Manufacturing Business Survey Committee. This index was at the lowest reading since June 1980 when it was 30.3 percent. In addition, Mr. Ore said, “New orders have contracted for 13 consecutive months, and are at the lowest level on record going back to January 1948.”

The new orders index was 22.7 percent in December, 5.2 percentage points lower than the 27.9 percent registered in November. No industry sector surveyed reported growth in December; the jobs sector particularly grim. The employment index was 29.9 percent in December, a decrease of 4.3 percentage points from November. That was the lowest reading since November 1982.

Emphasis added. The US index for new orders is at a sixty year low. Korea’s manufacturing output is shrinking faster than in the Asian crisis. Russian manufacturing is poised to shrink more rapidly than in 1998. China, Japan and Europe are all looking at manufacturing contractions too. JPMorgan’s global PMI is at — as one would expect — a record low. Edward Hugh reports that this should translate into a 10% plus contraction in global manufacturing output.

I guess macroeconomic volatility is not a historical relic after all.

My colleague Paul Swartz and the team at the CFR’s Center for Geoeconomic Studies have pulled together a set of charts to help track the current contraction by comparing current data to the average of past recessions. I personally found the charts useful — and I would be interested in your feedback as well. Are there indicators that we should be tracking that we aren’t? And are there indicators that we are tracking that aren’t that interesting? I am pretty sure Paul will be updating the charts regularly. Adding in the results of the latest ISM survey on Monday is an obvious first step …


  • Posted by modgen

    What is the definition of the new orders index? It is hard to believe that the dollar volume of orders is at a 60 year low.

    Is it akin to saying the “worst holiday season in 20 years” where retail sales have gone back to the level of a couple of years ago rather than growing?

  • Posted by pwm76

    Link to Paul’s charts is not working.

  • Posted by MiTurn

    Thank you for the graphs. They are very helpful in understanding the current economic environment. While I am hopeful, 2009 doesn’t see to promise an end to the recession until — fingers crossed — the 4th quarter.

    Great site, keep it up.

  • Posted by bsetser

    modgen — i am not 100% sure (I would need to look at the ISM site) but it certainly isn’t the $ value. It is meant to be a gauge of whether — relative to current levels of production — things are getting better or worse.

    pwm76 — could you give me more details. the link is working for me (I am using mozilla firefox) but i had trouble accessing it on the 31st from a different computer, so i believe that there is a problem. I am just not sure quite what/ where it is.

  • Posted by jonathan

    I wonder if instead of the silly argument against macro volatility we’re seeing a 1st recession in which global financial links cause a more rapid set of changes – which would increase volatility. If we go back to the mid-70’s, finance looked nothing like today and the world was then divided East-West with China on the sidelines. The connectedness of financial markets should mean more rapid handling of news & policy and this may then accentuate volatility. Will it then push us into worse, meaning that better financial information and average firms being closely tied to international finance markets will drive more radical cost savings, production decreases, etc. than if companies were still acting in a much more information constrained world? (Same for governments.) Or will this mean a hard, fast down and then a rapid up?

    One could construct a model on these terms but I don’t know how much real world meaning it would have.

  • Posted by Blissex

    As to indicators, I can’t remember where I saw it suggested (some other blog) but marginal GDP/debt ratio (how many dollars of GDP require a dollar of debt) is very interesting I’d think. Probably imports/debt would also be interesting.

    Also one that I saw a couple of times illustrated by Ritholtz, percentage of stocks purchased on margin.

    Then another I have seen is net borrowing as percent of disposable income…

    Another one that pops to mind is median wages (yearly) vs. both imports and exports.

    Another one would be estimates of net legal or illegal immigration and GDP.

  • Posted by Blissex

    What about percent of stock purchases or sales by non residents?

  • Posted by don

    So, we will see how my old prognosis turns out. Here and elsewhere I have opined that the recession would be substantially worse in China and Germany, as they depend on goods production and exports. So far, China’s imports have fallen more than apace with its exports, but this phenomenon has limits (besides any that come from international pressures), because China exports much more than it imports.
    The business cycles have abated in the U.S. over time as the economy has become more services-oriented and the old accelerator-multiplier has lost steam. But this mechanism may hit China and Germany with a vengeance, particularly as Germany exports capital goods equipment.

  • Posted by bill j

    What’s the impact of deflation on all this? I note that the index price component is very low, reflecting the collapse in commodity prices from August onwards.
    Presumably the PMIs are in current dollars, so I guess that the fall in the nominal value of orders could exaggerate the decline in production, as it fails to account for the cheapening cost of production?

  • Posted by Bernardo

    I haven’t seen house prices. Am I missing something?

    Regarding oil prices: a bit more of explanation may be required. “Few prices have larger geopolitical consequences”, ok, but this does not mean that it is relevant at all.

  • Posted by Yoda

    Treasury holders got violated these couple days, ouchi. More pain to come for treasury holders in comming weeks or months.

  • Posted by Yoda

    Next, GSE MBS holders will be violated.

  • Posted by Kafka

    Thanks, nice charts. I suggest following the TED spread quite closely, debt to nominal GDP adjusted for hedonics and imputations to the extent possible (using whatever deflator seems appropriate to account for various methods of computing CPI over the years), U6 unemployment, S&P movements adjusted for inflation (which has yielded a negative return over the last 10 years), relative currency rate changes of the major Countries and defense spending amongst the various Countries (as it is a relative negative drag on growth versus social expenditures especially in debt financed governments). P.S., I know I am a dope but I have always enjoyed your blog (especially your funds flow work).

  • Posted by bryce

    I second Blissex on the usefulness some sort of debt/GDP ratio. Total US debt stands in the range of ~320% of GDP, roughly double the range of 20 years ago.

    This expresses much about why this is not a normal recession.

    A global measure of world money supply/world GDP would also be of interest.

  • Posted by martin

    Consumer sentiment seems to be in major decline. As we go from bad to worse this is something that should not go un-noticed.

    An econonmy focused on service/consumption will need these numbers to not contract much further. If the consumer stops spending then we have much larger concern.

    My concern is the Obama stimulus maybe ineffective. After the stock market crash and horrid unemployment numbers (media headlines) have kicked in a “physcialogical impact” for many is inacted.

    Thus consumers will spend any such stimulus to pay down debt, or perhaps save. This would offset an Obama stimulus to spend spend spend in order to hold up our econonmy.

    What the world really needs as a proactive Chinese and Japanese consumer. How do we get there? Bright Marketing! Study of consumer behavior on the eastern front. This is what should be the focus of our multi-national corporations.

  • Posted by Tom Geraghty

    How about some measure(s) of business investment demand?

  • Posted by question

    If these charts/stats worsen over the coming 3-6 months, then is it not fair enough to conclude that the U.S. entered Depression in 2009?

  • Posted by Mike Hudachek


    I am an individual investor, who became familiar with your work through Prof Roubini’s website before you left. I have a keen interest in these graphs; as I believe you and Prof Roubini have a demonstrated, great track record in accurately predicting the world’s economy.

    My comments about the graphs are as follows:

    – Excellent information
    – I believe more and more that we should be comparing this recession to 1929. The similarities are scary both from a data perspective and events. Looking at averages gives you the perspective that this is very different than other recessions; however, comparing to a similar time in history may provide more insight as to what may come. I think both comparisons (avg and ’29) would be of great benefit for different reasons.


  • Posted by greg

    I’m always skeptical of averages. How many data sets are included in the average? Is there a way the charts can also capture the hi-lo values that make up the values with some sort of band?

    Very well compiled info. I look forward to updates.

  • Posted by Terry

    Concur with Mike: Excellent information!

    The charted data appear to reflect how much worse a recession the current one is than average. Most worrisome is that none of the recession measures yet show any turnaround. To me, that’s a sign we have a ways to go, probably another year and another -3% GDP contraction.

    Again, thanks for making this available.

  • Posted by Blissex

    «the usefulness some sort of debt/GDP ratio. Total US debt stands in the range of ~320% of GDP, roughly double the range of 20 years ago.»

    I have seen graphs over the past 30-50 years of that ratio.

    But what I would like to see is a slightly different thing: the ratio between changes in GDP and changes in debt. That it is to treat debt as a quasi-factor of production, and to look the “debt intensity” of the economy. The blogger mentioning this said that one dollar of extra debt has resulted in an ever smaller increase in GDP, and currently IIRC it is less than one dollar (e.g. the increase in debt is faster than that in GDP) and soon for the USA it will be negative, as a dollar more of debt results in a decrease in GDP, at which point very very bad things happen.

    As to that, long terms graphs of changes in GDP and in oil imports or total consumption are also interesting.

    BTW apart from BradS other sites have interesting long term series. Yahoo! Finance has got long term stock market prices, the St. Louis Fed has got lots of long term national and regional series, there are some very nice periodically updated long term graphs here:

    Mosler Economics has periodic posts of several graphs:

  • Posted by Twofish

    don: So, we will see how my old prognosis turns out. Here and elsewhere I have opined that the recession would be substantially worse in China and Germany, as they depend on goods production and exports.

    I’d argue that opposite which is that this recession is going to be less bad in China as it *doesn’t* depend on goods production and exports. If you look at the Chinese economy, it’s dominated by domestic infrastructure spending and investment, with exports playing an important but secondary role, whereas in the United States the economy is consumer driven.

    For China to boost demand, all that it takes is someone in Beijing to say “more railroads.” By contrast for the United States to boost demand it requires either an increase in consumer confidence or else a rather large restructuring of the US economy away from consumption and toward investment, much of it state driven.

    Well, we’ll see….

  • Posted by OGT

    Twofish- My understanding is that China has been intensely investing in Infrastructure for years. I think you overestimate how much room China has to increase this. Plus, China is at least as dependent on US consumer demand as we are, if not more, that’s where that trade surplus thingy comes from.

  • Posted by babar

    i like the graphs quite a bit but as usual it would be nice to see variance bars on the historical averages.

  • Posted by pwswartz

    The ISM number is very useful in that it is forward looking but it is also just a diffusion index; or a measurement of how broad the decline not directly how deep it is (although they usually very correlated). I’d be willing to meekly argue that the ISM number is implying a worse growth story than is happening; I suspect this because given the nature of the slowdown – indiscriminate credit contraction – I expect everything to slow, which makes the ISM number really ugly.

    Greg – I’ll agree that averages are only useful when viewed with the proper level of skepticism. The average lines will have a different number of recessions depending on the duration of the various time series; admittedly this is not ideal but one never has the data that one would like. I agree that having a measure of dispersion (high-low or st. dev) does have some value and if we can add that without making the charts too complicated for our average reader we may do that.

  • Posted by ckeating


    Mr Swartz’s charts are indeed very helpful. I think tho that one key one is missing – real estate values in previous recessions vs this one. That’s incredibly telling given that, as far as i know, only the ’91 recession saw a drop, a only a small and momentary one at that. All other post-WWII recessions have seen a rise in real estate values. This one is seeing deep and prolonged declines, which is, o/c so central to almost everything else

  • Posted by Flute

    Housing prices (e.g. Case-Shiller, preferrably inflation-adjusted) would perhaps also be useful to see, since housing has so far been a driving force in this recession.

    But very useful charts anyway – thank you very much. And thank you for a very good blog, which is on my must-read list.