Top-down versus bottom-up approaches' ability to anticipate turns in the economy

[myriam2222]

My understanding was that top-down approach was better at anticipating a change in the business cycle than bottom-up approach because it uses macroeconomic data, while bottom-up approach first works on single companies forecast which is in practice often based on historical forecast of the company concerned and sometimes tends to extrapolate a bit too much what has been achieved recently without taking a macroeconomic view of what could happen.

However i found various statements that are (at least I find) contradictory in this regard:

In question 3)A) of reading 16, the right answer is ”The top-down approach is less optimistic when the economy is heading into a recession than the bottom-up approach.” This is in line with my understanding because the top-down approach may foresee the recession that the bottom-up approach did not anticipate.

In Question 16) of reading 16 however, the answer states “Because the insurance company wants to detect quickly any significant turn in equity markets, Carmichael should recommend the bottom-up approach because the bottom-up approach can be effective in anticipating cyclical turning points.”

 

[cpk123]

In question 3)A) of reading 16, the right answer is ”The top-down approach is less optimistic when the economy is heading into a recession than the bottom-up approach.” This is in line with my understanding because the top-down approach may foresee the recession that the bottom-up approach did not anticipate.

Quote:
your interpretation here I believe is wrong.
The top down approach is LESS OPTIMISTIC than bottom up. So it detects recessions SLOWER than bottom up. Company (at the bottom up level) knows better than industry when recession comes faster.
I think the combination of the usage of double negatives - LESS OPTIMISTIC … and talking about Top Down has thrown you off in your interpretation. This statement simply put means - Bottom up detects recessions as well as up turns first. Because company has to do whatever it has to - only then does the information move up to the industry.

And this is consistent with Q 16 - where an insurance company (bottom up level) can detect quicker changes in the equity markets.

 

[myriam2222]

Well if bottom up is more optimistic than top-down heading into a recession, it means that bottom-up detects recession slower than top-down and not the other way around.

 

[cpk123]

Sorry. More optimistic - detects it faster…
You are looking at things optimistically - you look at the glass as half full. So more optimistic - detects faster.
why are you not thinking it through. Would a company know faster that things are going bad, or would the industry first realize the same and then it becomes evident to the company in that industry?

 

[vida_blue]

myriam2222 wrote:
My understanding was that top-down approach was better at anticipating a change in the business cycle than bottom-up approach because it uses macroeconomic data, while bottom-up approach first works on single companies forecast which is in practice often based on historical forecast of the company concerned and sometimes tends to extrapolate a bit too much what has been achieved recently without taking a macroeconomic view of what could happen.

However i found various statements that are (at least I find) contradictory in this regard:

In question 3)A) of reading 16, the right answer is ”The top-down approach is less optimistic when the economy is heading into a recession than the bottom-up approach.” This is in line with my understanding because the top-down approach may foresee the recession that the bottom-up approach did not anticipate.

In Question 16) of reading 16 however, the answer states “Because the insurance company wants to detect quickly any significant turn in equity markets, Carmichael should recommend the bottom-up approach because the bottom-up approach can be effective in anticipating cyclical turning points.”

The bottom-up approach may be better at predicting “black swan” type turning points. A recession after 5 to 7 years of strong economic growth isn’t hard to suspect by a top-down analyst. It’s a natural cyclical change and they have plenty of macro data to help predict it.
What those top-down analysts have trouble seeing is some sort of major shock to the sytem that comes seemingly out of left field. The bottom up guys may have a better chance of spotting this.
The classic example is Michael Burry going short the mortgage crisis. He wasn’t a macro analyst that was looking at economic trends and got worried about housing in Florida. Rather, he was looking at things on a security by security basis. He looked at the mortgage securitizations and saw massive, massive risk so went long CDSs.

 

[gad4]

I guess the answer is if we’re looking for ane economy/ recession type concerns we are doing top/down.
If we are looking for something that looks for a equity/change in a sector, we are doing more a bottom up?

 

[crosstheevil]

cpk123 wrote:
Sorry. More optimistic - detects it faster…
You are looking at things optimistically - you look at the glass as half full. So more optimistic - detects faster.
why are you not thinking it through. Would a company know faster that things are going bad, or would the industry first realize the same and then it becomes evident to the company in that industry?

It’s hard to wrap my head around this … Here is my 2c: a company can be an exception vs. the industry (think extremely good or bad), which explains why bottom-up is more optimistic coming into a recession and more pessimistic coming out of a recession. Detecting a recession is a different matter, because of its tendency to “overeact”, bottom-up “reacts” to a recession faster …

 

[myriam2222]

vida_blue wrote:
The bottom-up approach may be better at predicting “black swan” type turning points. A recession after 5 to 7 years of strong economic growth isn’t hard to suspect by a top-down analyst. It’s a natural cyclical change and they have plenty of macro data to help predict it.
What those top-down analysts have trouble seeing is some sort of major shock to the sytem that comes seemingly out of left field. The bottom up guys may have a better chance of spotting this.
The classic example is Michael Burry going short the mortgage crisis. He wasn’t a macro analyst that was looking at economic trends and got worried about housing in Florida. Rather, he was looking at things on a security by security basis. He looked at the mortgage securitizations and saw massive, massive risk so went long CDSs.

Thanks, yes, I get what you are saying. This seems like the most plausible explanation to me to explain why the curriculum says that bottom-up approach is better at anticipating turning points (because finally I understood that it’s the position taken by the curriculum).
For more standard turns in the economy (the “recession after 5 to 7 years of strong economic growth”) I would tend to think the top-down approach is better at anticipating, just like you mentioned and in line with what gad4 is saying as well. BUT that’s not the position taken in the curriculum.
At the end, I think the answer to the question “is top-down or bottom-up approach better at anticipating cyclical turns?” is quite subjective. In my business we do bottom-up and honestly I have seen many different people doing many different business plans and I don’t feel like anything great was done to anticipate cyclical turns.
But the curriculum takes a clear position on this and I think now we have to learn by heart that “ bottom-up is better at anticipating cycles”. That’s it….
Here’s a good and short discussion on this on the forum last year:
http://www.analystforum.com/forums/cfa-forums/cfa-level-iii-forum/91343034

 

[myriam2222]

crosstheevil wrote:
around this … Here is my 2c: a company can be an exception vs. the industry (think extremely good or bad), which explains why bottom-up is more optimistic coming into a recession and more pessimistic coming out of a recession. Detecting a recession is a different matter, because of its tendency to “overeact”, bottom-up “reacts” to a recession faster …

Yes exactly, I think the question on “more or less optimistic” was asked “considering everyone knows already that a recession is starting”. I.e. in that question it was not about figuring out which approach was better at anticipating the recession because the recession is a given fact (I understood that only after I launched this post).

 

[vida_blue]

myriam2222 wrote:
At the end, I think the answer to the question “is top-down or bottom-up approach better at anticipating cyclical turns?” is quite subjective. In my business we do bottom-up and honestly I have seen many different people doing many different business plans and I don’t feel like anything great was done to anticipate cyclical turns.

Agreed, it’s a pretty crude generalization but that’s the answer they wanna hear so I”m sticking to it