The Nightmare Scenario… “Unsafe” Havens

Thanks to policy-induced crowding-in of Treasuries, the front-end of the curve is now doing exactly what the Federal Reserve feared the most. A nightmare scenario of cutting rates… and nobody caring. Positive historical skew has made them a popular diversifying asset for the momentum crowd – and that’s exactly why we should be concerned.


The degree to which risk assets are beholden to easy policy has been well-documented. In addition, there is a natural inclination to be willing to exchange some risk of future pain for the promise of a positive carry portfolio. But, as a wise man once said: “the road to hell is paved with positive carry…” We’d be remiss to ignore that warning.

Here’s why.


Risk is commonly associated with the realized volatility of an asset. When you consider a “risk-neutral” allocation, that generally means people are expecting future realized volatility to roughly match the recent past. But that’s not quite right – and it’s the “not quite” that has formed the basis of the most spectacular blowups the market has witnessed.

That’s because we don’t just care about the amount of realized volatility. We also care about what direction our returns “lean” (i.e. skewness) & we care about our exposure to fat-tails, especially those to the downside (i.e. kurtosis).  

Consider the distribution of weekly returns for the S&P since 2001.

If you look closely, you can probably also see that the distribution appears to be “leaning” to the right. This is indicative of the fact large losses tend to happen more often than large gains. The “skewness”, as it’s termed, is -0.99 in this example. In other words, you frequently make a little (most weeks, in fact), but there are a few real disasters that can pose very real threats to the health of your portfolio. You’re exposed to large downside events.

Some of the brightest minds in the business have taken to task the idea of 100% equity portfolios ever since it was “proposed” by Thaler & Williamson (who… as you might guess… are not risk managers or traders).

My personal favorite rebuttal:–Equities

Consider a like distribution for the Nasdaq & the iShares MSCI Emerging Markets ETF.

Notice the same effect where the distribution leans to the right: again, negative skew at work.


It turns out this is the case in nearly every global equity market. Losses tend to be much bigger than gains, despite an overall upward trend for global stocks. This is a feature that’s been well-documented over the years. I’ve taken the below table from a piece co-authored by Jean-Phillipe Bouchaud (founder and Chairman of Capital Fund Management, professor of physics at École polytechnique & co-director of the CFM-Imperial Institute of Quantitative Finance at Imperial College London) entitled “Risk Premia: Asymmetric Tail Risks and Excess Returns”.

It’s also been the subject of numerous (excellent) blog posts on the topic, most notably in my opinion by Corey Hoffstein in “No Pain, No Premium”. If you haven’t already done so, I’d encourage you to set aside a little time for a close read:

An aside: interestingly, note that (despite a tendency towards negative skewness of the country’s bond market) Italy actually ranks at the top with positive overall skewness – a sign, perhaps, that 61 governments since WWII can rack up a lot of debt financed pro-cyclicality!


The problem of deeply negative skew is not true for every asset class, however. It’s less pronounced in commodities, though can still be observed to a degree.

It’s when you move to assets like bonds & yen FX crosses, that we start to creep out of negative territory.

For example, below I show the iShares ETFs (IEI & IEF). Fees are part of the reason these distributions are still slightly negative, but that’s a conversation for another day. Below I tag Treasuries by their point on the curve (UST = Treasury).


It’s only when you are willing to accept either lower mean returns or larger overall variance in these types of products where the skewness really starts to move positive.

The same can be found in certain FX crosses (especially those involving JPY) where there’s a tradeoff that’s made either between cost of borrow or average return.

[Note: the savvy investor probably knows there is a key “asset class” that I’ve left out. That’s momentum – which trades like a “long volatility” vehicle but with much better ex-ante Sharpe than just going out and buying every option in sight. It’s the one demonstrably reliable source of alpha that exists for what we can only term “behavioral” reasons. AQR again has an exceptional piece on the topic that is a must read for any serious investment professionals:]

You may also now be wondering how “risk parity” enters the equation. Well, it’s because they are achieving the best of both worlds by combining assets with a floating allocation that allows them to take advantage of the positive skew inherent in bonds & the attractive returns priced into equities.

Here is a toy distribution using just SPY & TLT. You can see that with just a few minor tweaks, the profile doesn’t look too bad at all.

These types of portfolios have done exceptionally well over the last 25 years, as evidenced by recently minted CT billionaires.

But they also mean that allocations can, and frequently will, shift heavily towards one side or another. Given the relatively attractive recent profile of Treasuries over the past few quarters, that exposure has likely grown.


Here’s a grid of what we’ve just considered – along with a few other important statistics (wkly data as of 2001), sorted by ex-ante Sharpe ratio.

Note the asset which creeps to the top of the list: front-end Treasuries.

The fact that the skewness is also at the top of the list is noteworthy. In other words, we have an asset which consistently delivers positive returns which are tilted towards larger gains for the winners than losers. That’s a pretty compelling reason to own the lowest yielding “risk free” instrument on the board – especially, and this is key…

…since inflation is virtually non-existent.


The catch? That’s in the column labeled “Kurtosis”: a measure of “fat-tailedness”.

Despite the fact 1-3y USTs have all the qualities listed above, it has a degree of “fat-tailedness” that’s significantly greater than other points on the Treasury curve – more on par with growth stocks than highly-rated sovereign bonds.

You are giving something up to get that positive skewness & high ex-ante Sharpe. You’re exposed to fat-tails, and despite the fact there’s a tendency for those to deliver in your favor, there will also be nasty shocks in the other direction.

In market terms, that’s a “carry washout trade”.

For more detail on how that affects the Treasury market – and why this represents a nightmare in the making for the Federal Reserve, see below.


And we’ve just gotten exactly that: a 25bp move higher in 2yr rates over the past 5 days which essentially eats away all the carry you would have expected to earn… for the rest of this year.

That’s without making any assumptions for financing or adjusting for inflation – which would make the outcome for the holder significantly worse!

Indeed, that’s also a record which stands going back over the last 10+ years for a move higher in yields. After we’d just set the 10-year record for a 5-day move lower in yields.


Now, that being said, there’s one retort that’s all too common these days. That’s the assumption that foreign investors will swoop in to buy our bonds on a mammoth backup in yields such as this.

That might be the case for dollar-reserve manager sovereign wealth funds (yes, that means China – but also a few others petrodollar nations like the UAE, Norway, Saudi Arabia, Kuwait, etc). They already have dollars on hand from exporting goods & thus have to spend those dollars somewhere…

And if this description of dollar-reserve mgrs sounds as though (from the headlines over the past year or so) we might be setting up for an enormous eff-ing problem, congrats – you’re now more economically literate than the vast majority of the policymaking elite.

So, if not the sovereign wealth funds – then maybe banks or other foreign buyers who want to own them because US yields are so much higher than theirs?


Doesn’t work that way, sadly. Enter the costly world of FX hedging.

Unfortunately, when you consider what the effects of currency hedges have done in the interim, the apparent “pickup” for foreigners is nil in this backup.

In fact, for the average Japanese investor they would LOSE money by buying our bonds & swapping out the FX basis (roughly -25bps, but arguably not as bad as it’s been in the past…).


Bottom line… They ain’t buying them. Unless things start to get a LOT cheaper.

So, the average US investor is going to be the one to buy this pullback. Which means if you planned to own them until maturity, you’d be settling in for a barely 1.5% total return. Small wonder the curve has inverted to such a degree.

The scary thing is – that might be the best option, amongst a sea of terrible alternatives.

References, Sources : Bloomberg, HFR Research, SocGen Hedge Fund Monitor, CME Group.


This material may contain indicative terms only, including but not limited to pricing levels. There is no representation that any transaction can or could have been effected at such terms or prices. Proposed terms and conditions are for discussion purposes only. Finalized terms and conditions are subject to further discussion and negotiation. OTC Derivatives Risk Disclosures: To understand clearly the terms and conditions of any OTC derivative transaction you may enter into, you should carefully review the terms of trade with your counterparty, including any related schedules, credit support documents, addenda and exhibits.

You should not enter into OTC derivative transactions unless you understand the terms of the transaction you are entering into as well as the nature and extent of your risk exposure. You should also be satisfied that the OTC derivative transaction is appropriate for you in light of your circumstances and financial condition. In addition, you may be requested to post margin or collateral to support written OTC derivatives at levels consistent with the internal policies of your respective counterparty. I may have positions in assets mentioned above. In addition, I may change those positions – and frequently do.

This post was written by me, the material is my own (except where sourced), and it expresses my own opinions. I am not receiving compensation for it & have no business relationship with any of the companies and/or organizations whose assets may be mentioned in this post.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.