We previously presented a piece by SocGen's Albert Edwards that claimed that there is nothing now but to sit back, relax, and watch as the US becomes another Japan, as asset prices tumble, gripped by the vortex of relentless deflation. Sure enough, the one biggest bear on Treasuries for the past year, Morgan Stanley, is quick to come out with a piece titled: "Are We Turning Japanese, We Don't Think So." Of course, with the 10 Year trading at the tightest level in years, the 2 Year at record tights, and the firm's all out bet on curve steepening an outright disaster, the question of just how much credibility the firm has left with clients is debatable. Below is Jim Caron's brief overview of why Edwards and all those who see a deflationary tide sweeping the US are wrong. Yet, in what seems a first, Morgan Stanley presents two possible trades for those with access to the CMS and swaption market, in the very off case, that deflation does ultimately win.
Morgan Stanley's rebuttal of the "Japan is coming" case:
There are many arguments that suggest the US is going the way of Japan, and while UST yield valuations may appear expensive, a regime shift has occurred and we should use the deflation experience of Japan as a guideline. We respect this point of view, and our colleagues in Japan provide some compelling charts.
In Exhibit 3 we show how the richening in the JGB 5y led to a significant flattening of the curve. Ultimately CPI turned negative and Japan did in fact move into a period of deflation. It makes sense for the 5y to outperform, as investors believed in a low rate and inflation regime for an extended period. Most money is managed 5 years and in, which thus makes the 5y point so attractive in low rate regimes because it represents the greatest opportunity for money managers to own duration, yield and return. The same is happening in the US as the 5y point is richening extensively as investors seem to be surrendering to a low rate and low return environment. But this may be premature.
Note that it took ~2-years before the JGB 2s10s curve started to flatten, predicated upon the richening of the JGB 5y on the 2s5s10s fly. And the extreme richening of the JGB 5y required CPI to turn negative (i.e., deflation). Thus, using Japan as a baseline, it may be premature to conclude that the richening of the UST 5y will quickly lead to a significant flattening of the UST 2s10s curve. Our colleagues in Japan are quick to point out that deflation and curve flattening in the US may happen faster because the US has the example of Japan to follow. However, we believe the US is far away from realizing negative CPI prints, and playing for a similar trade to Japan appears to be a low-quality bet.
Well, Morgan Stanley has been wrong long enough, perhaps it is their turn to shine now, and for the first time in history see an outcome that proves equity correct over credit. Of course, we won't hold our breath, however we are more concerned that Morgan Stanley's argument is more predicated by its purely bullish read on the economy, which in turn Caron believes should translate into far higher yields. Alas, he may have picked a wrong time for press with the bullish case: with Goldman having gone off the permabullish reservation, the trade will now be how to frontrun the other sellside strategists before they also go bearish. Alas, we think Caron is dead wrong by following the arguments of MS' economic strategists, the case may most certainly arise that absent the Fed getting involved in QE2, a scenario very much priced into the curve currently, that bond prices will indeed plummet if the Fed does not provide any of the much anticipated dovish disclosure at this Tuesday's meeting. We do agree that the long Treasury trade is the "consensus" trade now, and as always happens, the unwind of a groupthink trade is usually accompanied by blood, tears and toil.
Some more observations by Caron on why Treasuries are ridiculously overbought:
US Treasuries have become the default asset to purchase for those who expect a slow economic recovery absent of Fed rate hikes and for those looking to hedge systemic tail risks, the possibility of more QE, deflation or otherwise. We’re concerned about this because it may be becoming a consensus trade. And it runs counter to the evidence of improving macro risk conditions: i) less likelihood of a double-dip scenario, ii) a soft-landing in China and iii) reduced risk in European sovereigns. These are three pillars upon which our more positive macro outlook stands and why we think UST 10y yields can move toward the 3.25% area. Quite simply, we think investors are wrong to be so complacent about buying USTs at such expensive valuations. Here is what we mean:
Investors Are Not Being Paid Enough to Take the Risk of Owning USTs
The UST 5y has benefited most from the recent rally in bonds and is now trading at historically rich extremes (Exhibit 1). But the counterpoint is that the 5y also has the best roll down and carry on the curve. It is true that the expected return on buying the UST 5y, assuming unchanged conditions, is approximately 3.00%1. But the question is if a 3.00% return is enough to justify the risk of buying such historically extreme richness in the 5y point. We argue that it's not.
One should require a much greater payout ratio since a mere 74bps rise in yields could wipe out the 3.00% return. Even at today's low levels of vol, 1-year vol for the 5y is 97bps, which means 5y rates are priced to move up or down by +/- 97bps, implying that a 3.00% return provides only 0.76-STD (74bps expected return/97bps vol) of cushion if rates were to rise. If justification in owning the 5y is the pricing of a deflation scare, and client surveys show only a 15-20% probability for deflation, then investors should be looking at closer to 5:1 payouts. Otherwise owning the UST 5y at these levels is simply a low-quality bet.
Frankly, we do not think the investor is being paid enough to take the risk. These days, earning a 3% 1-year return on the UST 5y seems quite appealing. But as we mentioned in our piece last week, the easy money for the bond bulls has been made; initiating new purchases at today's low yields may prove to be more treacherous. In Exhibit 2 we illustrate UST returns since 1995 for perspective, which puts into context the unattractiveness of buying the UST 5y at these levels, holding it for 1 year, and earning a paltry 3% return. Such a return seems very low relative to the risk; unless of course we follow the path of Japan.
While we are sure Caron means well, the one thing he blatantly refuses to discuss is the elephant in the room: the Federal Reserve. It is glaringly obvious that should the Fed step in with round two of QE, regardless of how big it is, the Fed will most certainly purchase Treasuries. And if Goldman is correct, at least $1 trillion worth of them.
So for those who are more in tune with the Fed's reality distorting practices, and are willing to frontrun the Fed as Bernanke attempts to push record low yields ever further right on the curve, here are two trades recommended by Morgan Stanley to take advantage of the deflation trade.
While our core view is not for a US deflation, we recognize that investors may need to hedge against such a scenario. We recommend:
- Buy a 1y dual digital, which pays out 100bp in one year if 2y CMS is below 0.8% and 30y CMS is below 3.3% at expiry for 16.5bp. This trade offers a payout ratio of 6:1.
- 1y 1s5s conditional bull flattener, for zero cost, struck at 126bp. Currently, the spot 1s5s curve is at 130bp.
1y Dual Digital on 2y CMS and 30y CMS, with Strikes at 0.8% and 3.3% We recommend a 1y dual digital on 2y CMS and 30y CMS, which pays out 100bp in one year if 2y CMS is below 0.8% and 30y CMS is below 3.3% at expiry, for 16.5bp.
Currently, spot 2y rates are at 0.71%, and spot 30y rates are at 3.76%. This dual digital offers a payout ratio of 6:1 if deflation occurs in the US.
This structure allows investors to take advantage of three favorable relationships, which
cheapen up the cost of the option:
- Rolldown: Currently, spot 2y rates are at 0.71%. In one year’s time, they are priced to increase by 60bp to 1.31%. The investor is cheapening up the cost of this option by moving the strike of the 2y leg 51bp OTM relative to the forward, but nevertheless 9bp higher than the spot 2y rate. Similarly, we target 30y CMS because they have the least amount of rolldown on the curve.
- Cheap skew on 2y tails: Implied volatility on low strikes for 2y tails is lower than implied vol on ATM strikes. This works in the investor’s favor.
- Cheap volatility on 30y tails: Currently, USD 1y30y volatility is cheap versus the VIX, versus FX implied volatility and versus EUR 1y30y volatility (Exhibits 1-3)
The risk to this trade is that 2y rates in 1 year are above 0.8% or that 30y rates in 1 year are above 3.3%. If either of these occurs, then the investor’s downside is limited to the initial upfront premium of 16.5bp.
1y 1s5s Conditional Bull Flattener for Zero Cost
Investors may also hedge a US deflation through a 1y 1s5s conditional bull flattener for zero cost. This trade expresses the view that the 1s5s curve will flatten, but only if rates rally.
Specifically, we recommend:
- Buy $100mm 1y5y receivers struck 23bp OTM
- Sell $478mm 1y1y receivers struck ATM
- Net cost of the trade is 0bp upfront
The recent flattening of the very front end of the US curve has caused the carry dynamics of front-end steepeners to change. One year ago, the 1y 1s5s curve was 80bp flatter than spot. Now, however, the 1y 1s5s curve is 19bp steeper than the spot 1s5s curve, and as a result, 1y 1s5s flatteners roll positively (Exhibit 4).
In this trade, investors are selling 1y1y volatility (at approximately 78bp norm), and buying 1y5y volatility (at approximately 98bp norm). The ratio of these volatilities works against the investor, which is why the 1y5y receiver that investors are buying needs to be moved 23bp OTM in order to make the trade zero cost.
Net, this means that the entry level for the bull steepener is at 126bp, which is only 4bp flatter than spot. Exhibit 5 shows the historical level of the 1s5s curve versus the entry level for the conditional bull steepener.
For investors to profit on this trade, 5y rates in one year’s time need to have rallied by 23bp more than 1y rates. This is with respect to the current forward rates. Today, spot 1y rates are at 0.48%, spot 5y rates are at 1.74%, 1y1y rates are at 0.95% and 1y5y rates are at 2.4%.
While an outright 1y 1s5s flattener has positive rolldown, investors lose if the curve bull steepens or if it bear steepens. By expressing this view conditionally, investors are only exposed to bullish scenarios at expiry. In other words, if at expiry, the 1s5s curve has bear-flattened or bear-steepened (relative to the current forwards), then both options expire OTM, and the investor neither gains nor loses money. Investors gain if the 1s5s curve bull flattens, and lose if it bull steepens. In the latter case, downside is potentially unlimited.
With a key decision due out of the US central banks this week, the wait for whether Morgan Stanley is finally correct will not be too long. In the meantime, cheap deflationary trades like the ones presented are likely the most profitable bets for the time being. We likely have at least a few more major Fed interventions before the hyperfinlationary collapse predicted by Edwards is reality.