Sunday, April 8, 2018

Lending and the business cycle

Since I think tight mortgage lending has been the key variable holding back more healthy economic expansion, I have been watching some bank lending measures for signals of contraction.  I also think monetary policy is too tight because the Fed is using inflation measures that include a large amount of measured inflation that is due to the non-cash effect of owner-occupied rent inflation.  That high rent inflation is also a result of tight lending, which is keeping housing supply low.

There is an interesting relationship between commercial and industrial lending and the yield curve.

Lending increases during an expansion, with a bit of a lag, so it has parallel movement over time with the yield curve (here, 10 year minus 2 year, inverted).  Commercial and industrial loans tend to grow the most when the yield curve is relatively flat.  Then, when the yield curve inverts, this seems to be correlated with economic contraction and a newly steep yield curve.

But, this relationship has changed recently.  The yield curve remained relatively steep even while C&I lending moved back to expansionary rates of growth.  I suppose this would commonly be explained with the incorrect idea that the Fed has been holding short term rates too low, and that has been the source of the steep yield curve.

I think the more accurate explanation is that when interest rates are low, the zero lower bound causes the yield curve slope to increase, because long term rates begin to act like at the money call options.  There is a sort of premium embedded in future rates.  So, from 2008-2012, the relationship is similar to the typical relationship, but the yield curve is unusually high (lower on the inverted chart).

Until QE3, lending was starting to slow down, and long term rates were declining because economic sentiment was declining.  QE3 turned that around.  The yield curve steepened, and with a bit of a lag, C&I loan growth recovered.

Since then, it appears as if the relationship has reversed.  The yield curve has flattened while C&I loan growth has declined.  The first part of this period is from the end of QE3.  QE3 had boosted growth, so the end of QE3 was somewhat contractionary, but by the end of QE3, the neutral short term interest rate had probably recovered to something close to 0%, whereas it had previously been negative.

Then, the Fed began to raise short term rates.  This has two contrary effects on the yield curve.  The first effect is that it pushes the base rate up from the zero lower bound, which causes the inflated effect on long term rates to decline.  In other words, long term rates start acting more like in the money call options, so the embedded premium is lower.  This flattens the yield curve.

To the extent that the Fed raises rates back above the neutral rate, this also will flatten the yield curve by pulling down economic sentiment.  Both of these things are happening at the same time, which is doubly flattening the yield curve.  One effect is caused by expansionary factors (a rising neutral rate) and the other effect is caused by contractionary factors (a rising policy rate).  So, the question is, what effect is most at work.

And, the fact that C&I lending growth has been declining suggests that monetary tightening has been too strong.  There are many economic measures that seem to be at fairly benign levels, so this hasn't been confirmed by other data, so I have been in sort of wait-and-see mode.  Inflation data also hasn't confirmed either thesis, as both backward looking (non-shelter) inflation and forward looking inflation have been fairly flat, even if backward looking non-shelter inflation has been quite low.

I have had a bias for expecting the bear-reading of this situation to eventually play out, but recent lending has only complicated this, because suddenly, there has been a little bit of a spike in lending, and real estate lending has very slowly been growing for a little while already.  Is this a turnaround to new recovery, or is this a liquidity grab by corporations because nominal revenues are starting to weaken?  There isn't much evidence that things are bad enough to expect that to happen, so I have to wonder if this is part of a turnaround.

The mystery continues.


  1. I wonder what fraction of C&I loans are collateralized by property?

    In other words, property loans in drag.

    That source has been tapped out?

    Great post.

    Maybe we are stuck in New normal: slow growth low inflation no recessions....

  2. - You should look at "change in debt". As soon as the growth of the change in debt starts to go down (e.g. from say +2% down to say +1%) in a country then that country is already in recession.

    Source: Steve Keen

    (I would love to post using my DISQUS profile. Can this be added ?)

    1. Thanks. Do you have non-video links to this idea?

      Do you know how to add Disqus on blogger? There are several profile sources that show up for me, but Disqus isn't one of them.

  3. - Watch the 1st video from about the 30th minute. From there on Steve Keen provides some info this concept of "change in debt".
    - Steve Keen has a stunning simple formula:

    Income + change in debt = aggregate demand

    Multiply that formula with "change in" then the formula becomes

    change in (income) + change in (change in debt) = change in (aggregate demand)

    Now assume (for simplicity sake) that "change in (income)" is zero then the formula becomes:

    Change in (change in debt) = change in (aggregate demand)

    This last formula has a devastating message: To get a rising aggregate demand with income remaining flat one needs to accelerate the GROWTH of debt.

    - Take the 1st formula (income + change in debt = aggegrate demand) and use some concrete figures:

    Let's assume that income is at 1000 and the change in debt is +200, then the formula becomes:

    1000 + 200 = 1200 (aggregate demand = 1200)

    Now assume that income remains at 1000 and "change in debt" still grows/is still positive but at a lower pace, say +100 then the formula becomes:

    1000 + 100 = 1100

    So, demand drops from 1200 down to 1100, merely because the growth of debt drops from +200 down to +100. And shrinking demand leads to rising unemployment.

    - About DISQUS: To be honest, I haven't got the faintest clue. I use DISQUS for a number of other websites. Haven't got any of the other profiles. Sorry mate !!!

    1. Interesting. That makes sense.

      I think one of the problems that has infected discourse about the bubble is that, regarding this relationship, monetary policy has been implicitly assumed to be fixed. To the contrary, an inflation targeting central bank will naturally adjust for this. So, there are many who argue that households used their housing ATM's to overconsume. But, that was simply a nominal issue, and the Fed pulled back currency growth to contradict it. It wasn't a shift to overconsumption so much as a substitution of households substituting debt for cash. Then, that debt growth collapsed in early 2008, and the Fed didn't take the extreme actions that would have been requied to counteract it in the other direction. If they had, they would have been accused of pumping up the alleged overconsumption.

      In the end, I'm not so sure there isn't some causation in the other direction - that slow growth in currency led to more growth in debt.

  4. With over 10 years of experience helping people acquire, recover and stay in their homes and expand/start their business.
    We help new entrepreneur and smart business owners to grow with adequate finance.

    Ford Credit Centre offers business loans to help you grow your business.
    We give out all kind of loan like Educational loan, Business loan, home loan, Agricultural loan, Personal loan, auto loan and other good Reason, I also give out loans from the rang of $5,000USD- $5,000,000USD at a 3% interest rate. Duration of 1- 15 years depending on the amount you need as loan.

    Get back to us for more information

    WhatsApp:+1 404 400 4210