On Sunday, using an analysis by Deutsche Bank's Dominic Konstam, we showed why the Fed by miscalculating the equilibrium real rate (which in reality is negative due to the massive debt overhang and the declining nominal growth rate, as well as record deflation exported by every other developed nation and now, China) is about to unleash a major policy error by hiking rates in 3 weeks - a process which will promptly result in a recession and ultimately force the Fed to both cut rates to zero or negative, and proceed with even more QE.
Here is Konstam's conclusion:
This is the important policy error scenario because even a very shallow path of rate hikes might drive the real Funds rate well above the short-term equilibrium real rate, further depressing demand. It is then plausible that the economy would be driven into recession, and the Fed would quickly be forced to abort the hiking cycle. As an aside, such a policy error could reinforce itself by causing structural damage that puts additional downward pressure on the equilibrium real rate. In this case the yield curve would flatten meaningfully, at least until the Fed actually reversed course by cutting rates.
* * *
This scenario is also bullish for rates because the Fed would, at the very least, stop rolling down its SOMA portfolio. More likely it would restart asset purchases in an attempt to stimulate the economy once more, pushing yields further down. We have argued in the past that unconventional forms of monetary accommodation are here to stay. The minutes of the October meeting confirm this view, noting that some policymakers felt it would be “prudent to have additional policy tools” because a lower long-run equilibrium real rate makes it more likely that reductions in the Funds rate alone would not be sufficient to stimulate the economy in the event of a downturn in the future.
Whether or not Konstam is right that the Fed is about to make what will prove to be a credibility-crushing mistake, remains to be seen. However, for those who, like us, believe that anything the Fed does is one mistake after another and thus a December rate hike will merely be the latest one in a long series of policy errors, the question is: how should one trade ahead of this error.
The answer comes from another research analyst at Deutsche Bank, Aleksandar Kocic. Here is his complete trade cocktail in three parts for the next 12-18 months that will encapsulate the distinct phases of the upcoming market tailpsin: the bear flattening of the curve, the bear steepening, one which may see the 10Y trade at or below the 2Y, in the process flattening the curve and perhaps even leading to the dreaded inversion.
From Deutsche Bank:
The year ahead: Policy mistake in three acts
In our view, the next 12-18 months will be divided into three periods corresponding to the three distinct regimes of market dynamics. They can be summarized by the following modes of the curve: short-term tactical bear flatteners on the back of a Fed liftoff story, followed by volatile bear steepeners of the “taper-tantrum” type around mid-year, and a bull-flattening finale as structural factors deem rate hikes to be a policy mistake.
Stage one: Bear flatteners
In the next 3-4 months Fed liftoff should be the main theme, with tactical bear flattening revolving around its timing and intensity. The Fed is likely to hike in December, and market prices are already reflecting this. However, conviction about a March follow-up is lower, although in our view this is a real possibility. This phase is bullish for USD and, given the expected slow rates rise, rates vol is likely to be relatively low and concentrated in the upper-left corner as the short end of the curve remains sensitive to the data, while the long end continues to be range bound.
As Fed liftoff represents the moment of decoupling between different economies, currency play will remain the main mode of adjustment to that reality. Therefore, although reversal of monetary policy is transmitted through rates, volatility will migrate from rates to FX and possibly risk assets, depending on how risk reacts to it. To some extent, we have already seen vol market positioning for this transfer of vol. Figs 1 and 2 show the recent history of rates, FX and equity vol.
In general, one should keep in mind that hikes are practically self-defeating as they lead to a stronger USD, which could slow the pace of a rate rise. So, a March hike is considered somewhat of a tail risk. We are buyers of leveraged bearish expressions in the ULC:
- Buy $100mn 4M2Y 25bp OTM payers subject to KO if 2s exceed ATMF+25bp in the first two months, 5bps offer vs. 13bp for the vanilla (62% discount)
The discounting is a function of both delta and serial correlations. A vanilla articulation in terms of payer calendar spreads would consist of financing a 4M2Y payer position with 2M2Y. For comparison, we adjust the strikes so that the premium matches the exotic trade:
- Sell $100mn 2M2Y 20bp OTM payers vs. buy $100mn 4M2Y 25bp OTM payers, offer 5c
In the vanilla version we have to sell 2M2Y 20bp OTM payers to finance 25bp OTM 4Y2Y payers. The difference between the two trades is their risk profile – while the exotics trade has limited downside (max loss = options premium), the calendar spread is vulnerable to MTM losses in case of transient spikes in gamma. Losses could be theoretically unlimited, although this is an unlikely possibility.
Stage Two: Bear steepeners
There have been many attempts to infer the Fed reaction function post-2008. Given the structural changes and breakdown of standard macroeconomic models in recent years, the question remains regarding the metric that captures economic recovery and the exact monetary policy response to it. The most easily understood indication of structural changes is the drop in GDP and the output gap relative to the potential that has opened up since the early days of the crisis. The growth has been so slow that based on the extrapolation of the pre-crisis potential, it was hard to see any point on the horizon when the gap would close. However, accounting for all the changes that took place, the revision of the potential was the key to actual convergence – while growth was increasing at a moderate pace, potential growth was even slower. This is shown in Fig 3.
According to the most recent data, their intersection is anticipated sometime next year. We believe that, regardless of the Fed reaction function, the market will interpret this as a bullish sign for the economy (and bearish for rates). At that moment, it is likely to challenge the Fed’s gradualism and begin to price the Fed as being behind the curve. This is the bear steepener, taper-tantrum type of mode.
The main mode of repricing is likely to be a selloff in long forwards and revisions of the terminal Fed funds rate, in particular 5Y5Y. 10Y UST yields could reach or even exceed 2.75%. We would position for this phase through quiet bears in mid-curve:
- Sell 26bp OTM 6M 5Y5Y MC receivers vs. buy 6M 5Y5Y MC ATMF/26bp payer spreads
The trade has a max upside of 26bp running and is vulnerable to rally 26bp below forward with theoretically unlimited downside.
An alternative expression of this mode could be through 6M 2s/10s or 5s/30s curve caps, or OTM conditional bear steepeners. However, while these are attractive positive carry trades, given the expectations of short-term bear flatteners and potentially volatile repricing in the belly, we would wait for better entry levels in Q2 next year.
A better articulation of the bear-steepening trade in the context of the outlined sequence of curve modes would be in terms of curve-cap calendars: Sell shortdated caps vs. buy long-dated OTM caps:
- Sell $1bn 3M 2s/10s ATMF (105.5bp) curve cap vs. buy $1bn 6M 2s/10s ATMF+15bp (113.5bp), costless
The trade has a (slight) positive carry (favorable time decay from short gamma partially offset by curve rolldown). MTM risk could be theoretically unlimited in case of transient short-term stepeners.
This period is bearish for USD and possibly for risk assets. It is likely to be very volatile. Vol should return to rates and migrate from short to the back end of the curve. Equity and/or credit vol could also rise significantly if bear steepeners persist and threaten to cause redemptions and forced an unwind of the bond trade. We would be buyers of forward vol at the back end of the curve along the lines of 6M 3M10Y FVA.
- Stage three: Bull flatteners
Despite everything that could happen in the near term, the main long-run story remains lack of demand. With all else equal, the market can decide to interpret rates selloff and Fed liftoff as policy mistakes and price in the adverse impact on growth and position for further rate cuts in the future. This is the third phase, the twist of the curve – the front end remains constrained by the Fed, while the back end rallies. It is a bull flattener, after a continued rise in rates. We anticipate the 10Y UST yield to rally towards 2% after trading as high as 2.75% during the second phase. This is generally bearish for USD and for risk assets, and as such could mean higher equity vol. Given that rates vol should reach high levels during the bear-steepening phase, bull flatteners would bring back vol sellers and return of the carry trade.
To capture the path from high to low rates, we would be buyers of knock-in receivers:
- Buy $100mn 1Y10Y ATMF receivers subject to KI if 10s touch 2.65%, offer 80c, a 73% discount to vanilla at 300c
The trade has limited downside with max loss equal to the options premium.
In general, one can capture the entire path of the second and third stages through mid-curve vs. vanilla switches. The logic is that of mean reversion: If rates initially do one thing, they should subsequently do the opposite. To capture the yoyo mode in rates, we recommend the vanilla/mid-curve payer switches:
- Buy $100mn 18M10Y vanilla payers at 365bp vs. sell $100mn 6M 1Y10Y MC payers at 205bp for a net 160bp
This is a positive carry synthetic forward vol trade (the payer side). It is an articulation of mean reversion. If rates sell off in the first 6M, MC payers are in the money, so the position consists of receiving fixed on 1Y10Y swaps (selling a payer option) and being long 1Y10Y vanilla payers. From put/call parity, the net position is a swaption receiver. The trade is short gamma and is vulnerable to transient spikes in vol in the first 6M with theoretically unlimited MTM losses.