Let’s back up and check that global macroeconomics 101 holds true.
The theory is that, base case, countries forecast to have high GDP growth --> end up having high GDP actuals --> which drives high corporate earnings --> and stock prices increase faster than low GDP growth countries assuming multiples hold constant. Even if DM multiples expand (as they are now), and EM multiples contract, this mispricing can not last forever, therefore over the long-term high growth countries should outperform. So let’s just see…
So there you have it, not complicated. Dividends vary by country and time, but will not change this result. Currency risk is overblown, currencies are relative, if you are USD based you take massive gains on RMB (+28%) and massive losses on BRL (-32%), and these tend to offset over time in a long-term diversified portfolio. “Risk adjusted returns” are made up, that’s just shit DM people say to justify their underperformance.
So I guess we’re really talking about “short-termism”, information bias, and nationalistic bias. Base case, today’s countries forecast to have high GDP, will have higher returns by 2025, yet people want to chase silly stuff rather than long-term outperform.
That is correlation, not causation. GDP refers to production within a country’s borders, so if DMCo is selling to EMppl, it wouldn’t show up in DM GDP.
Not disagreeing with you but you’re ignoring 3 vital components here…
Dividend reinvestment on the S&P500 adds +50% return over the last 10-years
The Rupee has depreciated by 40% since 2005, brazil is a full fledge currency crisis
Inflation in India, Brazil and China is 4x as high as US, Japan & UK - India alone was at 10% annualized inflation during 2011
No doubt China is a front-runner for future equity investment but you can’t just craft optics that make returns look sexy without consideration of the facts.
Palatir, why wouldn’t DMco selling to EMppl show up in DM GDP? Wouldn’t it show up in the X factor in:
Y = C + I + G + (X - M)
On the other hand, PA is wrong in that DM *equity growth* from EM sales are not baked into DM *GDP growth* in any systematic way. Export revenues are, but the growth in equity that results from it is not (though if you use the Economists “stock building” factor, then you can say that it is, but it’s a nearly trivial sized correction, and even tinier when you realize that stockbuilding is the entire equity valuation, and not just the EM part).
It also gets trickier with all those foreign profits that aren’t repatriated? Do they show up in earnings and retained earnings because they have been earned, or are they excluded because they haven’t been repatriated. My guess is that they are included, but it would suggest two different valuations depending on whether those things get taxed or stay abroad, which has nothing to do with company operational performance or GDP growth.
GMO has a lot of good research on the relationship between long-term GDP and Equity returns and has changed my thinking on this topic. That said, earlier this year they decided to abandon the US and get more exposure in EM. This is because medium term GDP forecasts are useful in predicting equity returns – to the extent they are accurate…
If you look at data, as opposed to stories, long term GDP growth historically show a negative correlation with long term equity returns in both DM and EM cases. This suggests that long term GDP forecasts - even if they are accurate - are at best irrelevant and at worst counterpredictive of equity returns. Ben Inker of GMO has a good paper on this that uncovers why this is: there were a number of reasons, and the only one I remember right now is that - although the growth does happen - equity holders get diluted by new investment that’s required to obtain this growth.
Say AAPL factories in China sell iPhones to India. The U.S. isnt exporting here, China is. If that iPhone is fold in China, it is an internal transfer.
Certainly in the short and medium term, change in valuations makes a bigger difference that total growth, but in the long term, you would think that valuations would cycle around an average multiple, so that when you look at long term index growth, the trend would be some multiple of underlying earnings growth and the valuation changes would just average out into via periods of optimism and panic.
But it doesn’t really work out that way, and part of the reason is dilution via new share issuance.
don’t forget inflation. high gdp growth comes with high inflation and inflation is equities’ biggest nightmare as it most often comes with higher interest rates which makes FI more attractive and raises the cost of capital across the board.
I think the inflation effects are mixed (though admit I’d have to look at the data again to sift out more): Real GDP growth does not necessarily come with high inflation, though it is true that high inflation will push up nominal GDP growth, which is why you might see GDP and inflation figures linked. And it is also true that in EMs, high GDP growth has often led to high inflation, but this seems to be a story that was more common from the 1950s to the 1980s than from 1990 onward.
Or were you talking about the Philips curve, where generating higher than average GDP would seem to imply higher than average inflation rates. My sense is that the Phillips curve is more of a short term thing and not a long-run thing, just reminding people that you can’t stimulate to eternity without inflationary consequences (though it doesn’t appear to have held true in the latest cycle).
Inflation is not necessarily bad for stocks, because required returns have an inflaiton component to them and so present values get discounted for expected inflation and the returns will tend to track inflation + a premium provided the anticipated inflation is accurate. However inflation *surprises* (when they are above expected values) are definitely bad for stocks, because they push up the required returns, and lowers their present value. Nonetheless, over time, that should wash out, unless you are in a hyperinflationary situation, in which case prediction becomes nearly impossible.
on an obseravational basis only. it is lesser so about inflation directly and more about interest rates. higher gdp growth tends to result in higher wage growth which tends to result in higher inflation which tends to result in higher interest rates. it’s my understanding that higher than average inflation and the resulting higher interest rates is why higher levels of GDP growth are bad for equities. i thought it was more of a funds flow issue, not only seeing money flowing into bonds instead of equities but also money moving from spending to saving (savings rates are higher when rates are higher), though a tangible rise in the cost of capital should have an affect on equity valuation also.
Why are “DM people” always trying to manipulate the numbers? CN CSI-300 has higher dividends than the US S&P500 (2.4% vs 1.9%). If you calculate with reinvested dividends for the US, obviously you do so for CN and other countries also. Cmon people.
Exactly, this is a very basic principle everyone should understand, and is what I am trying to communicate on this thread. The “PA formula” should be taught in CFA – “for a long-term holding period, starting from fair valuation, and ending with fair valuation, the country with higher earnings growth (likely related of higher GDP growth) outperforms…thus buy high-growth low-historic-valuation and sell low-growth high-historic-valuation”. Obviously this is a starting place for global macro analysis, not an absolute trading rule.
Sure US/JP stretches to 22X right now, and years ago CN crashed to 8X, but over time there is an avg multiple, and highs and lows tend to pull back to this average. You can’t beat forever on multiples expansion, eventually you get kicked in the face by the fundamentals.
What we have today are people chasing DM multiple expansion, when the fundamentals (GDP growth and earnings) just aren’t there. For example, as of yesterday trailing 12 month as reported earnings for the S&P500 were -11.5% vs one year prior , yet the index rallies and is now 22.15X. Notice how nobody is talking about that -11.5% number? That’s right out of the S&P Excel Download. Can’t igore that forever.
I think this goes back to my earlier point that these ETFs seek to track an underlying EEM Index. BUT, that index is garbage and does not actually represent the real emerging market securities, in their real global weights. For example, if they hold CN they always hold H-shares (as is the case with VEIEX which tracks the FTSE Emerging Index), yet H-shares have massively underperformed A-shares over the last 10 years.
I actually do not know of a single emerging, or global equity index which is properly constructed, as these have all been constructed by DM workers. See MSCI ACWI global equities, total junk. This is sort of annoying as I have nothing to benchmark myself against, I easily destroy the crappy EM benches, way too easy and I feel I am cheating.
BTW I advoce country-picking and stock-picking, not a fan of going long EM as a category via an ETF.
I would like to know that exact number but I don’t have a Bloomberg terminal and don’t know of any source that does apples to apples comparison. For the last 10 years obviously the CSI300 beats the S&P500 on price return, dividends, and currency. There is no mathematical way out of that. If that were not true, given their huge growth rate over the last 10 years, something would be seriously wrong with valuations.
What other sources are out there that would have this information? There’s a suprisingly small amount of information relating to chinese equities available to US investors.
More curious than anything as it can only enhance the return %s
It’s super annoying. Beijing says they want to open the capital account and get foreign investors, but foreign investors require basic stuff like total return and earnings.
I’ve actually sent some emails, one last week to China Securities Index Co aksing why they do not publish basic things like what S&P500 publishes – EPS by quarter so we can get greater understanding of what is happening to P/E, total return instead of just price return, ITD returns. It’s super stupid, nowhere on the freakin’ internet can I find China earnings, nowhere. Backing into them via P/E tells me they are positive, so why not show that off?
It’s just bad marketing. Christ if they paid me I could make their monthly fact sheet a work of art, they have massive 10yr outperformace yet they only show to 5yr (which underperforms). Are these people idiots? Of course no answer to my polite emails asking for data or at least clarification on existing data which is poorly labeled. Here is the fact sheet, not much facts…need a $20K USD Bloomberg Terminal.
