positive “shift” factor of yield curve

Reading 24 Yield Curve Strategies Page 203: “The positive (negative) “shift” factor is a non-parallel increase (decrease) in all yields.”

Is it parallel increase or non-parallel increase? Recall from Level 2, it seems to be parallel increase.

I have the same question: upper on that page they even explain that shift means a non-paralel change in the level of interest rates, which I’m not able to place as a non-paralel change for me is either a slope change or change in the curvature.

I assumed there are 3 possible scenarios of YC change:

  • paralel shift (down or up) labelled change in level

  • slope change (steepening or flattening)

  • curvature change (less curvature, more curvature - measured with the butterfly)

What is non-parallel shift then???

Change in curvature is a non-parallel shift.

As is a slope change.

Well, as per curriculum it seems not. Change in curvature, change in slope and “non-parallel shift” are 3 different scenarios.

I have no new curriculum so cannot check.

Parallel change is shifting entire yield curve up or down driven by one single factor.

Term structure is differently affected along its regions by IR changes which causes non parallel yield curve changes which is, in my opinion, more common situation in reality.

I suggest you to draw on paper likely situations with parallel and non parallel shifts for changes in yield.

As I can recall, the change in slope is stepping or flattening and change in curvature is twisting. Parallel change is simply shifting a yield curve in parallel fashion up or down. There are not 3 different kinds of shift beside a parallel and non-parallel. However, within non- parallel there are more scenarios.

Parallel shift is measured with ordinary duration. Shifts inside different regions are measured with Key rate duration.

This is what curriculum writes (as it was already cited by OP above):

“The positive (negative) “shift” factor is a non-parallel increase (decrease) in all yields. A positive (negative) “twist” is a flattening (steepening) of the curve, while in a positive (negative) “butterfly,” the two ends of the curve move downward (upward) and the middle of the curve moves upward (downward). These three empirically derived movements correspond well with the more stylized movements emphasized in the earlier discussion of trading strategies”

(Institute 203)

Institute, CFA. 2019 CFA Program Curriculum Level III Volume 4. CFA Institute, 5/2018. VitalBook file.

The citation provided is a guideline. Please check each citation for accuracy before use.

Have bolded and underlined the expression in question. Seems as either an error (have not found in errata though) or we (OP and I) misunderstand sth.

In other words, the positive (negative) “shift” factor would differently affect segments (aka regions) of a yield curve. E.g. 5 Y yield would be increased for 10 bps while 10 Y yield would be increased with 5 bps.

“FACTOR” is the key word maybe. I think I understand based on your explanation, Flashback!

Very unfortunate paragraph of the curriculum for a non-native speaker.

Factor is one positive or negative shock which affects a yield curve but differently along its term structure.

The influence of such shock would affect a curve by shifting in a non-parallel fashion (e.g. twisting) rather than in parallel shift.

I would recommend additional literature once you pass this exam. I’m planning to purchase this one very soon:

Thanks very much Flashback!