High vs. Full market segmentation

F

FrankCFA

New member
Joined
Jun 18, 2026
Messages
0
Reaction score
0
Is below statement correct? Thanks.
- High market segmentation => diversification advantage; correlation between market is low
- Ful market segmentation => no diversification advantage; correlation is 1 (relevant global portfolio is the individual market).
 
From what I have understood (and somebody else please confirm, correct or extend my logic) it sounds right:
If markets are fully integrated, then capital flows freely between them and hence you can use a correlation value.
If markets are fully segmented, there are some barriers to moving capital between markets and correlation is 1. Perhaps someone can elaborate on this last point as well? If a market is fully segmented, why would its correlation with the global market be = 1? i.e. it moves hand in hand with the global market? Wouldnt it be more isolated with a very low correlation instead?
 
Thanks arigolden. I have the same query as you.
 
arigolden wrote:
If a market is fully segmented, why would its correlation with the global market be = 1? i.e. it moves hand in hand with the global market? Wouldnt it be more isolated with a very low correlation instead?
Click to expand...
The justifications and examples (eg North Korea) I’ve read so far were too lame and ambiguous at best. It’s nonsense to say that the correlation between Canada and the US is closer to one (before Justin Bieber) than the correlation between France and the US. And still when it comes to North Korea, we consider the correlation to be 1.
Someone has mentioned that with full segregation, the segregated market is your full universe. If this is the case then why calculate its characteristics in an international portfolio context.
 
So you’re facing the same quintessential CFA curriculum confusion as us. I hope someone is able to provide us with a more coherent answer as this seems like an important and highly testable area.
 
Keep in mind the stinger model is a ICAPM risk-premium approach for calculating a risk-premium for markets that are essentially difficult to invest in and therefore deserve a risk premium. This isn’t a exact science - nothing in finance is, it’s just an approach/guide.
It’s not wether or not the market is 100% segmented or 100% integrated, what matters is the degree of such that determines the necessary risk premium over the global market.
 
Thanks Galli but that still does not address why the correlation for a fully segmented market and the global market is equal to 1. I’m sure we can memorize this but it would help us all if we can understand why
 
A fully segmented market assumes the global market is actually itself, therefore it’s a correlation of 1
They will give you a correlation for the fully integrated view which scales down the fully segmented expected return. The degree of integration will determine the weight of the fully integrated risk premium + (1- w) for the fully segmented risk premium
The two added together is the total expected return. I realize how confusing this is, just do 20 practice problems and it’ll click eventually.
 
For Singer and Terhaar analysis,
- the more illiquid an asset is, the greater the liquidity risk premium should be.
- the more segmentation an asset is, the greater the segementation risk premium should be. <- Correct?
 
suspense wrote:
suspense wrote:Is ICAPM in the 2015 curriculum
Click to expand...
No answer from the forum
Click to expand...
ICAMP is crucial!
ICAPM:
R_i =R_f +β* (R_m –R_f)
R_i – R_f
= β* (R_m –R_f), β = Cov/ (σ_m)^2
= Cov/(σ_m)* [Sharpe Ratio_m]
= ρ_i,m *σ_i * [Sharpe Ratio_m]
 
You must log in or register to reply here.
Back
Top