TN_Boy wrote: ↑
Sat Oct 05, 2019 4:38 pm
Returns for international are affected by exchange rate fluctuations. Which I, apparently incorrectly(?), call either exchange rate risk or currency risk. Though I'm not the only one using those terms interchangeably: ...
I sorta see what you are getting at, but it seems to be complicating the picture. I'm looking at nominal returns in US dollars and I sorta thought that was what all these intl/US performance comparison were doing that. Separately I might look at US inflation to get real returns.
First, I misread your post I though you were confligating FX movements with FX risk. You were not and I apologize.
Second, it does not complicate the big picture, rather the opposite.
We care about real returns, not nominal returns. Thus we should disregard the inflation component of returns.
Equities tend to be a good hedge against inflation.
FX rate movement = Change in interest rates + change in relative strength of the currency
Most FX movement is tied to inflation.
Thus FX risk is low.
This holds pretty well in practice. Over the past 100 years the annualized volatility of Inflation Adjusted FX Risk
is 1%. As Vineviz points out, FX risk is bounded by Purchasing Power Parity and thus means revert - one of the few risks that actually does. That means the longer you hold intentional stock, the lower the risk. If you hold intentional stock for 10 years FX risk tends to become noise. There may be one exception, which has been the last 10 years, where the USD has gone from a historically weak position a historically strong position. I would really like to see some decent work here.
vineviz wrote: ↑
Fri Oct 04, 2019 10:15 pm
The easiest way to decompose this, I think, is to look at three MSCI indexes:
1) MSCI USA
2) MSCI ACWI ex USA in USD
3) MSCI ACWI ex USA in local currency.
The difference between 1 and 2 is the total return differential for a US investor. The difference between 2 and 3 is the effect of exchange rates, and the difference between 3 and 1 is the difference due to underlying performance (e.g. sector composition, systemic risks, idiosyncratic risk, etc).
From 1988 to present, 1 minus 2 has been roughly 5%/year. 2 minus 3 has been roughly 3.5%/year and 3 minus 1 has been roughly 1.5%. So that's how I got the 2/3 attribution for currency effects.
Now I tend to think that PPP holds over the long run, so expect to see mean reversion in exchange rates.
Vineviz, I really respect you and I think you get things right more often than I but you are off base here. There is a difference between FX movement and FX risk. I believe that America has had higher inflation over the past 30 years and that we can ignore. I think this is where the bulk of the difference is coming from. Where FX risk comes in is the unexpected relative strengthening / weakening of the dollar. This is low (1% volatility) and bounded (PPP). I just have a hard time that a bounded mean reverting risk parameter
Next, I think you are off on " 3 and 1 is the difference due to underlying performance (e.g. sector composition, systemic risks, idiosyncratic risk, etc)." "MSCI ACWI ex USA in local currency" is in the multiple local currencies so it would also included all FX changes.
I have read far and wide on this, but I will point to the MSCI index calculations whitepaper which uses FX rate changes and does not try to account for the relative change in power of a currency. Plus Financial Market History: Reflections on the Past for Investors Today by David Chambers et. al. Plus The Economist reporting on its Big Mac Index. And a key article in the Financial Analysts Journal.
That being said, I know this is a tough nut to crack. We are talking about relative changes between FX rates, so there are no external or absolute points of reference. While PPP and CPI are the best measurements we have, they are still pretty rough measures. I would be appreciative if you could point to any solid articles out there.