Kevin M wrote:Thanks siamond. Any quick update on the HUGE discrepancies in international stock returns?
Let me detail where the numbers are coming from in both cases, and you'll be able to judge by yourself which series you prefer...
Simba data sources:
- Total International is approximated to EAFE (i.e. Developed/International) for 1985-1987, as we don't have any reliable numbers for EM in this time frame (a topic I'll elaborate on within a few days).
- EAFE is in turn coming from the MSCI EAFE NR USD data series.
- Previous versions of Simba were using the MSCI EAFE
GR USD data series, but all Vanguard International funds track the 'NR' indices, so I fixed this in my working version, using 'NR' series in all cases. Differences are fairly minor.
- and then yes, the 1985 return (before any ER adjustment math) is 56.16, while 1986 goes to a dizzying 69.44 and 24.63 in 1987. What a good time for an International investor, according to MSCI.
For Portfolio Visualizer, it is a little more complicated to explain, but interesting, so bear with me:
- Let's start by the 'Quality minus Junk' monthly data series found
here on the AQR Web site (told you, bear with me!)
- Download the equities data set (add aggregates if you wish), and look at the MKT tab. This is source data for the 'quality minus junk' computation (which we really don't care about), let's stick with the source data.
- Those are International returns for various countries (from Austria to Sweden and USA), and one aggregate is 'Global ex USA'.
- Ah, but wait, there is a subtlety. pvguy told me that those are excess returns over risk-free returns (e.g. US T-Bills), and he's correct (see below).
- So do some math to add the T-Bills returns and convert the monthly returns to annual, and you will find what Portfolio Visualizer tells you, I just checked with 1985, 1986, 1987. And yes, 1985 is strikingly different, around 15.3.
Ok, let's dig one step further:
- the 'quality minus junk' data series came from
this article from Cliff Asness and al.
- this paper indicates "
Our sample consists of 39,308 stocks covering 24 countries between June 1951 and December 2012. The 24 markets in our sample correspond to the countries belonging to the MSCI World Developed Index as of December 31, 2012. We report summary statistics in Table I. Stock returns and accounting data are from the union of the CRSP tape and the XpressFeed Global database. All returns are in USD, do not include any currency hedging, and are measured as excess returns above the U.S. Treasury bill rate."
- a few sentences later, it says "
As shown in Table I, with the exception of Canada (whose coverage starts in 1982) for most countries XpressFeed’s Global coverage starts in 1986. Our sample runs from January 1986 to December 2012.". Er, what?
- hm, let's come to the AQR spreadsheet, and pay closer attention to the year 1985. Ah yes, Global Ex-USA is indeed equal to Canada! Whoah! All Canadian posters on this forum suddenly feel quite empowered, right?
You tell me, did I make my case?