Emerging Markets - 29/01/2014

So the CBRT had an emergency meeting where the following headline came through

*TURKEY'S CENTRAL BANK RAISES OVERNIGHT LENDING RATE TO 12.00%

*TURKEY'S CENTRAL BANK RAISES BENCHMARK REPO RATE TO 10.00%

So Turkey raised rates quite considerably and a full list can be seen here

CBRT rate hikes, HSBC

As expected the TRY caught quite a substantial bid, moving from 2.25 to around 2.15, not too bad. However there was one overarching problem with this, the substantial rate hike which accounted to about 225bps would severely damage the macroeconomic picture. That is on top of an already slowing economy so it really won't end well, but they had little choice... FX reserves are under $40bn and trying to talk the TRY higher failed many times.

So what what remembered was that this was the CBRT's last move, they couldn't realistically put rates up again, and couldn't really intervene directly. As such the TRY really struggled giving back all the days gains to trade back towards pre-CBRT levels

USDTRY 2 day tick

The market somewhat settled down, with USDTRY trading around 2.24 (stopping out HSBC after just 12 hours)

@UlasAkincilar @katie_martin_FX pic.twitter.com/JvLjhDhVKr stop was 2.24, now price touched that some places(on reuters), others came very clos
— Jeremy W-Smith (@JeremyWS) January 29, 2014

At this point the days move was already incredible, but what came next was utterly ridiculous.

*SOUTH AFRICA RAISES BENCHMARK RATE TO 5.50% FROM 5.00%

The market was a tad surprised to say the least with 25 of 25 economists expecting no rate hike. This however (much like BoE in 1992) made traders interpret the recent hikes as weakness on the part of both ZAR and TRY which both immediately started to tank.

USDZAR

The USDZAR reversed the initial spike lower on higher rates and proceeded to move 3% upwards, and over 4.5% from the days low (post CBRT)

This fear spread to all the asset classes and quickly we saw the USDTRY trade up to 2.32, causing risk off across the market, ES dropped, Bond were bid, JPY was bid etc etc.

ES white, TRYUSD yellow

However, more concerning was the contagion that spread in EMFX where the RUB weakened to 5 year lows vs. the USD and all time lows vs. the EUR

USDRUB 5yr

Not just RUB, but even the stronger (CA surplus EMs) ccy's sold off with the HUF down 1.5% vs. the EUR and PLN down 0.5%

All things considered, I really don't see this action today being a solution to the problem, it won't fix the structural deficit for turkey, it won't encourage Speculators to buy TRY (and could like with BoE 1992, cause the market to test the central bank and destroy the ccy) and finally it will hurt the macroeconomic picture for TRY.

In summary, today was the largest ranged day for over 5 years with around 7% from high to low, based on spot. It also pushed the 1 month IV to its highest since 2009

USDTRY range

TRY 1 month Implied Vol

Best part is, the FOMC haven't even annouced their plans yet, which will likely involve a $10bn taper, which could worsen the situation for EM FX even more.

Forward Rates

Since the beginning of 2014 there has been a significance divergence in some of the forward rates for major currencies. This should be a key factor going forward for the FX markets as the guidance and expectation from Central banks should be a critical factor.

Below shows some 1y rate 1y forwards

1Y1Y rates for GBP, USD, EUR, CAD

As we can see here, CAD 1Y1Y rates have dropped by around 25bps, EUR rates down by 8 or so. Yet USD and GBP forward rates have risen marginally.

This divergence will be very important when looking to the medium term, as we can the fixed income markets start to react and so far, in FX, only the CAD has responded to the changes in central bank outlooks.

Since the taper though, the market has moved to from a state where it was considering the (at the time) present factors, and instead has moved to considering how these are going to change over the next 1/3/6/13 months etc.

This point can be illustrated by simply looking at the EURGBP, right off the bat there should be no obvious implications of a Fed Taper, there is no excess correlation with the USD between either of the EUR or GBP and as such it should have been a neutral event. But it wasn't.

EURGBP hourly

As we can below - the EURGBP has traded mostly in tandem with 1Y implied yields (from forwards) for the most part of the last 5 years. But these 1Y rates on show the current picture, when looking ahead now, we need to consider how they rates are going to change.

EURGBP 1Y implied yield vs EURGBP

That is why, since the Fed Taper - and pretty much since it became an idea in the early part of 2013, the EURGBP has been trading much closer to that of the 1Y1Y spread. (shown below)

1Y1Y spread vs EURGBP (red) and 50 day correlation

We can see the correlation moved from mixed to decisively red, where red shows a negative correlation (because the spread is inverted on the right axis)

As we can see that now looking at 1Y1Y rates are far more important than just 1Y rates as now investors are caring much more about the central bank path than pre-taper.

The reason why the taper is so important is it has acted as a psychological change from "uber-dovish" to "cautiously hawkish" and this means that the markets will start to consider the fact that base rates will soon be on the rise (as opposed to the idea that we've ZIRP ad infinitum).

Now in the case of the EURGBP the central bank divergence is pretty obvious - ECB have recently cut rates and will have to ease more, where as the day-by-day pressure on the BoE to hike is increasing (consider U/E recently). As such selling rallies in EURGBP should be good all year.

Another example is AUDNZD.

Here is a chart of the AUDNZD spot, AU-NZ 2 year spread, and the spread between the 2 year rate 1 year forward.

AUDNZD vs rates

Once again, since mid 2013, the AUDNZD has become far more correlated with the forward rate spread as opposed to the current spread. The divergence between the RBA and RBNZ is quite clear, but relatively minimal in comparison to others.

So, looking forward, using the idea that the market will start, albeit slowly, to noticing forward rates more and more the EURUSD looks to be a good sell going forward. So long as the rates factor becomes more important than the relative balance sheet and also the strong Current account position of the EZ.

But providing it does, the EURUSD will come under heavy pressure as the expectations widen more and more, and the yield spread widens to a point where the EURUSD can no longer justly trade at these levels.

Finally it would be easier to just pay USD rates and recieve EUR rates but it should filter through to the FX markets.

And here where the current implied yield is in Blue as well, we can see how it is very much like EURGBP pre 2013 where it only cared about now, as opposed to the future.

This is could be what we are looking at if the EURUSD starts to price in the future as opposed to the present. (ok maybe not 1.20, but under 1.30)

AUDNZD downside vol too expensive

Firstly, the AUDNZD has been very, very weak over the past year, unsurprisingly to a point, but now it seems a little extreme to me.

Below we can see the AUDNZD plotted against the AU 2 - NZ 2 year yield spread, and an accompanying model (AUDNZD=Bid,0= 1.202 + 0.1465 x (AU2YT=RRBid Yield,0-NZ2YT=RRBid Yield,0)) to show the divergence when related to price.

AUDNZD, rate spread (red) and model (purple)

Right now, with spot trading 1.07, we are clearly below where the rate spread would otherwise see us trade. over the course of the year the AUDNZD and rate spread has an R^2 of 85% suggesting that the theory the related FX to rates held strongly.

However right now, with the 2 year spread at 41 bps this prices in the AUDNZD at ~1.14 (+/- 1.5%), but it is clearly outside of this range by quite a lot. But when using this relationship to see why we are trading where we are, we see that the AUDNZD is pricing in a rate spread close to 100bps (more precisely 90bps +/-8), more than 50bps wider than current.

AUDNZD model with 90 bps and 125bps 2 year spreads as question

As we can see here, at 90 bps the model expects the AUDNZD to be trading at 1.0702 (pretty much spot)

Which in essence given the relative similarity of the AUD and NZD yield curves (in terms of shape) and given their equal base rates, means that the AUDNZD is trading currently with ~50 bps of RBNZ rate hikes built in. With the RBNZ unlikely to raise rates 50 bps in the coming 6 months then we have some potential opportunities. First comes from the idea, that if the RBNZ puts of the inevitable rates hikes a little longer, then Spot should, over time move towards the rate fair value in which case buying spot at 1.07 would be a good idea, and even if the RBNZ does hike, it currently prices in 50bps as we trade now so there is a clear buffer.

RBNZ OCR rate path

Furthermore though, when we consider the volatility smile for AUDNZD we can build some ideas.

1M, 3M, 6M vol smiles

We can see that, somewhat unsurprisingly there is a higher IV for puts vs. calls. And this would make sense in the 1Y duration as that is most likely when the rate spread between AU and NZ is at its widest. Above in the calculation, 125bps is put in to the calculation and returns a value of 1.02 for AUDNZD. So a move towards parity is technically possible, but for now, a long way off.

But for now, when considering the 6M duration, the 10 delta put is trading at 8.21 (mid) and the strike on the 10D put is around 1.01

AUDNZD and 10 delta 6M strikes

So we can see that the 10D strike is 1.01, and given that the market is already pricing in 50bps of hikes, it is unlikely that in 6M that it could trade 600 points lower and price in another 35-40 bps of hikes. And therefore, based on the level of IV and the probability that it manages to trade that low, it would make sense to consider selling these puts naked. The heightened level of IV for the puts on the short term makes this trade, at least from a risk:reward standing very attractive to me.

A way to increase the premium (and in turn, the amount you receive by selling) would be by also selling the 10D calls with a strike of 1.16. AUDNZD is unlikely to recover that far in 6m due to similarity in the RBA and RBNZ rate paths.

So entering a 6M 10D strangle would receive approx. 57 pips or just over 0.5%, not huge, but when factoring in the probability of a move outside of those bands in 6M it is a very strong chance this trade will perform.

conversely you could increase the delta ( bring the strikes closer to ATM) and hope to receive more in premium and you would, but also decrease the probability quite substantially that the AUDNZD trades outside the band. For a 20 delta 6M strangle you receive around 129 pips (1.25%) but the strikes are 1.03 and 1.13.

But like I said, this isn't about being in a strangle as much as it is about the downside puts being far too expensive up to 6 months, based on the idea that the AUDNZD is unlikely to move to a situation where it prices in a 125bps differential on the 2yr.