Shorting EM

Hello, I am new to this forum. I am a retail investor. My hope is that I can explain my views, and maybe some of you can help me understand how I can most efficiently express my views as a retail investor.

I think there are massive problems in China. China represents 70% of global GDP growth. They are highly levered, but they have been able to “sweep” the bad loans under the rug. Because I think their leaders are quite cunning, and they have much more capacity to influence markets, I would like to bet against the periphery. I want to bet against those smaller economies that are tethered to the Chinese economy. These smaller economies do not have the currency reserves needed to combat this gobal currency war. Rates will inevitably move up in these countries and growth will be hard to come by.

I am also long bitcoin, a position I started building in May 2015. The top 20 countries whereby new BTC wallets were opened in 2015 were those in EM. When I saw this list, it confirmed that there is serious contagion building within the economic system. How can I efficiently short EM?


Well, the most obvious way is to sell short emerging market ETFs. Due to borrow costs, it would probably be most efficient to use the most liquid ETFs that provide the exposure you want. Also, I don’t know how specific you need to be with respect to asset classes (currency, government debt, equities, etc.), but for emerging markets, these tend to have positive correlation with one another.

You could also use options, but trading costs probably make ETFs more worthwhile.

I recommend shorting EWZ for Brazil. Worked well for me over the summer.

As ohai said you really cant get too niche, youll probably have to stick to the main country indexes or larger EM ETFs if you want to keep borrow costs managable.

I was looking at going long EUM, but even with 1x inverse, the price decay destroys returns over time. I suppose I could short EEM, but then I have to eat that 2.5% div yield. For instance, EWZ the yield is 4.2%. Did you eat that?

The dividend yield is not relevant - if a $100 stock pays a $1 dividend and nothing else changes, the share price becomes $99 on the ex-date. The implicit funding or borrow cost is relevant, though.

I know we were all taught this, but there’s really no way I’ve found to see it happen in practice since nothing happens in a vacuum. Just wondering if anyone really pays attention to this or if it’s just academic, much like the DDM. In theory, it’s absolutely correct. In practice, not so much.

I guess maybe seeing how a certain company, say Ford, does on its ex-date compared to its peers might make some sense, but I still don’t feel like that would yield any concrete data.

This is off-topic, by the way. Not talking about any ETFs or stocks in particular. Well, I guess not entirely off-topic since shorts would be disproportionately punished compared to the beneift of being long on the ex-date.

^the difference gets arbed pretty quickly

I use “EDZ” : 3X short MSCI EM ETF. I would not buy in at these levels though, I’d wait for a relief rally then buy once things have rallied back up. At these prices now and following this dip you’re just asking to get burned by running with the herd. Using derivatives isn’t practical and neither is traditional shorting for a retail investor. The ETF"s are the best way to go.

Noticeable for companies with double digit yields

I agree with you StL for large cap companies paying a small dividend. When coke pays it’s dividend there are so many other factors at play that the price is going to adjust based on them and not the “in a vacuum” value. This has been at the heart of my curiosity about dividends vs buybacks: the actual effect vs the text book effect

It may be a wash in the long run. Academically, you have to pay $1 for a $1 dividend if you are short, but the price is supposed to drop by $1 when the stock goes ex-dividend. So on the ex dividend date, you have $1 of profit from the short, plus $1 of loss from having to pay the dividend, which nets out to zero.

In the real world, you do get charged for the dividend (which gets paid to whoever lent you the shares), and the price generally drops by a comparable amount on the ex-divdend date, but there’s also noise from other stuff that is going on, so it doesn’t seem like the math adds up as perfectly as on the whiteboard.

But if you didn’t have a dividend paying stock, there’d be that same noise anyway, so I’m not sure this is a case where the academic version is so far from the reality. After all, when the stock goes ex dividend, it really is worth less than it was the day before; it’s not some regression prediction from a model with a bunch of assumptions that says that on average it’s worth less: you own the stock plus the dividend on one day, and just the stock the next. It’s some of the most basic math you can get in finance.

elton & gruber (1970) -

you must remember that from L2?

Yeah, ok, you got me. Taxes suck.

Thought it’s interesting to use an academic paper to bolster the point that the basic approach is “too academic.”

The main point still stands that the fact that you are on the hook for dividends in the short position should not be a major consideration in the cost of funding unless you are seriously illiquid (why would you be seriously illiquid anyway, if the stocks can be called away or short squeezed). If a dividend is 3%, then the net cost of a taxable dividend at 35% is just under 1%, which would be the predicted tax effect if all investors in the stock were taxable. You shouldn’t be shorting if 1% is likely to make a difference between profit and loss, and the difference is likely to be less than that anyway.

Thanks Dr. bchad. I wouldn’t have expected any other response from an academic. But that’s not where I’m hung up on the issue. I know the math works. Still unconvinced anyone really takes it into consideration when buying and selling the stock on the ex-date (though the point above about companies with large yields makes sense). It’s not like the stock automatically starts a dollar lower. Humans still need to price it in*.

*Though the algos probably do and since they’re taking over the world, the dividend payout could be priced in more efficiently now more than ever. I may have just come full circle.

Do you have any concrete data that shows that shorts are “disproportionately punished” on ex-dividend dates?

Oviously not. In fact, I believe I used those exact words, if not entirely in your exact context. But, thinking about it intuitively, shorts have to pay the dividend out of their own pocket. Money leaves their account/portfolio. We assume this is completely offset by the corresponding price decline on the ex-date. They’re exposed to a certain type of risk…not sure what it would be called, “dividend equalization risk?”

On the other hand, being long a stock you’re given money in hand and you still can assume your stock is going to go down by the corresponding amount, and if it doesn’t (and really it would only be an upside deviation) you get a bit extra.

My exact point is, shorts have to assume the market works perfectly and exactly the way it should. Longs don’t have to care and, in fact, benefit if markets don’t.

I don’t have the faintest idea how one would or even could prove this.

If anybody in a financial market is disproportionately punished, the empirical evidence should be there.

When you buy bonds, you pay accumulated interest, until the coupon resets. Why would stocks be different, at least between the announcement and the ex-dividend date?

Yes, there’s noise at the open which means it doesn’t trade exactly a dollar lower. And some people don’t pay attention to the ex-dates.

Sure, humans have to price it in, and maybe 90% of it happens on the ex date and the remaining 10% happens over the week. The point is that if you are concerned about having to pay the dividend the dividend while shorting a dividend paying stock, it will basically be a wash where your ability to be liquid when the dividend is due is pretty much the only consideration.

Now, there are lots of signalling things that might happen when the dividend is announced and/or paid, and those things can get into feedback loops that do wacky things to how the price evolves. No one’s denying that. But that doesn’t mean that the (stock + $1 dividend) = (same stock the dividend going to the guy who owned it yesterday) .

^Yeah, I’m not saying there’s a material impact. Just got me thinking. If the shorts were really getting screwed, market neutral products just wouldn’t work.