Brad Setser

Follow the Money

Cross border flows, with a bit of macroeconomics

Print Print Cite Cite
Style: MLA APA Chicago Close


More on the fall in private borrowing and the rise in the fiscal defict

by Brad Setser
June 2, 2009

The chart that supported my previous post attracted a fair bit of attention. But it was cobbled together at home and only looked at data over the past few years. With help from the CFR’s Paul Swartz, I looked at the quarterly data over a longer time period.

The story is clear. Government borrowing has increased dramatically. It topped 15% of GDP in the last two quarters of 2008. In 2007 and early 2008 it was more like 3% of GDP. But private borrowing has fallen equally sharply. Total borrowing by households and firms fell from over 15% of GDP in late 2007 to a negative 1% of GDP in q4 2008.


Negative borrowing by households and firms means, I think, that households paid down their debts in the fourth quarter.

It hardly needs to be noted that the fall in borrowing by households and firms in late 2008 was exceptionally rapid. A stronger economic cycle implied that the magnitude of counter-cyclical fiscal policy also needed to be ramped up.

The disaggregated data on borrowing by households and firms is also interesting. Household borrowing rose to record levels in 2003 and remained high through early 2006. Household borrowing fell in 2007, but for a time this fall was offset by a rise in borrowing by private firms. Borrowing by firms actually peaked in the middle of 2007 at a higher level than during the dot come investment boom. Chalk that up to a surge in leveraged buyouts and stock buybacks.


Both charts highlight the risk that worries me the most. In both the early 1980s and the first part of this decade, both the private sector and the government were large borrowers. And in both cases, borrowing rose faster than domestic savings, so the gap was filled by borrowing from the rest of the world. The trade and current account deficit rose. In the early 1980s, the US attracted inflows by offering high yields on its bonds. More recently, it did so by borrowing heavily from Asian central banks, together with the governments of the oil-exporting countries. But now yields are low (even after the recent rise in the yield on the ten year Treasury bond), and need to be low to support a still weak US economy. And China (and others) are visibly uncomfortable with their dollar exposure; banking on their continued willingness to finance a large external deficit seems like a stretch.

The challenge this time around consequently will be to bring down the government’s borrowing as private borrowing resumes.

Update: Household, household and government and total borrowing over time on a rolling four quarter basis:


And the quarterly data.


All data comes from table F1 of the flow of funds.

The downturn in total borrowing in the rolling four quarter sum reflects a low level of total borrowing in the second quarter (i.e. before government borrowing ramped up both to fund the fiscal deficit and to finance the financial sector). But even in the fourth quarter of 2008, total borrowing wasn’t up …


  • Posted by Winslow R.

    Krugman wrote “Where’s the money coming from?”

    From Krugman’s title I thought you finally had figured out how the monetary system works and where money comes from.

    Given how smart Paul is, I just don’t believe he doesn’t know. Though from what he writes, he’s not telling anyone.

    You’ve posted some pretty good stuff so was just hoping you did too and were actually spreading the word.

    No one needs to borrow from China.
    A properly functioning Treasury/Fed can peg any interest rate it desires as long as there is the political will.
    The political will is largely determined by the exchange rate of the dollar and what will help our banksters.

  • Posted by Jim Caserta

    I guess I”m late because comments 1&2 captured my sentiment. A graph of the sum would be a dagger through the heart of the argument.

  • Posted by anon
  • Posted by anon

    from your immediately previous post:

    “As households and firms rediscover their animal spirits and start to borrow more, the amount the government borrows needs to fall. Otherwise, total borrowing would rise, and the amount that the US needs to borrow from the world would rise.”

    This looks quite wrong based on the data series you are following.

    These are credit market series; not net surplus or deficit series.

    The total net borrowing series from F1 is far more volatile than the capital surplus series. This is because these credit market series are incomplete pictures of the entire flow of funds. There’s much more to net out in a complete analysis.

    Your graph is more coincidence than complete analysis. Interesting that Krugman used this based on your graph.

  • Posted by bsetser

    jim — i thought i had added those graphs, but i guess they aren’t showing up

    anon — you are right, these are credit market series (though they capture more than “market” borrowing i think, also bank lending), not surplus or deficit series. Not sure though where that takes you …

  • Posted by bsetser

    anon — someone in the past suggested that you need to aggregate the funding cos and the broker dealers, which sort of makes sense …

  • Posted by anon

    “Not sure though where that takes you”

    Just to point out its not totally accurate to associate the difference in government and private sector credit series with net capital inflows – although you make a qualification about equity not being included, there’s a lot more than equity that’s not included:

    e.g. of difference in range of items:
    Have a look at F.107
    Compare line 20 (credit component within ROW net financial asset acquisition) with line 13 (total ROW net financial asset acquisition)
    huge difference in size
    similar distinction for liabilities

    materially, you may be capturing the important relationship, but will that always be the case … ?

  • Posted by Ted Kaehler

    Please add in venture capital. Perhaps the dot com bubble will appear as a total borrowing peak,and the housing bubble as another.
    Perhaps the Fed should set a target of (government borrowing + personal + business + venture capital) = a fixed percentage of GDP. The government could lower its borrowing and raise interest rates when a bubble is forming.

  • Posted by davidbaer

    Having been a part of the Online Universal Work Marketing team for 4 months now, I’m thankful for my fellow team members who have patiently shown me the ropes along the way and made me feel welcome