Thursday, February 18, 2010

More on Leading Economic Indicators

dblwyo comments:

Jake - have you got data access to the LEI? I don't but Northern Trust has been
publishing charts for a while showing YoY LEI vs. GDP and those are very good.
LEI per se has not held up well on a MtM basis but the YoY comps show them
shining (and yes the subliminal Nicholson pun is intended).
Thanks to Dr. Donald J Oswald we all have access to LEI data.

While I recognize the LEI has led the recovery (that is the intention as it's "leading"), it has shown far more growth than the actual economy.

Likely because all the stimulus feeds directly into the components of LEI, but have not been reflected in the "actual" economy.

Source: BEA / CSUB.EDU


  1. Thank you kind sir - excellent. Leave us however no fall prey to the quantwonks syndrome. The last couple of years LEI may have been a tad "wonky" because the relationships have been estimated on normal business cycles. Right now we're in a different regime (phase-space) where a lot is being policy-driven. So stock and money indicators would tend to be mis-leading, at least imho. A discussion that hearkens back I believe to one of your first LEI posts.

  2. it was discussed at econompic here:

    basically, the current print turns negative if you disclude the steepness of the yield curve.

    as always... the world is an interesting place and it will be fun to see how it plays out

  3. So, M'sieu de Sade how does the whip feel now?

    Ran that data, converting everything to quarterly basis. Fascinating - two axis chart especially. Conincidence is remarkable BUT... there's a big divergence in this whole last decade that got worse. Didn't they give a Nobel to the guy who figured out how to estimate non-stationary time series? :)

    BtW - just to check an QtQ vs. YoY on LEI and there's no gain in signal but a lot more noise. Not sure about monthlies but on the whole the GDP is as much it's own leading indicator as the LEI is. You wanna tell 'em or ... ?

  4. Seems like there's a pretty strong historical precedent for peak LEI = peak GDP growth for a given cycle...

    ...and it seems like the risks of LEI falling are increasing: it looks like the the supplier output bump we saw in January won't be replicated in February, money supply has nowhere to go but down, building permits unlikely to show any strong gains, and unwinding of the "Euro fear" trade will put upward pressure on short-term rates

    Need a sharp decrease in initial claims or increase in confidence/consumer order to keep this gravy train going

  5. Jake, do a regraph with the new 2s10 spread after the FED rate hike LOL!