Goldman Slashes Treasury Yield Forecasts
If it appears like it was only yesterday that Goldman was advising clients to short the 10 Year Treasury, it is because it was... give or take a few months: From January: "Since the end of last August, we have argued that 10-yr US Treasury yields would not be able to sustain levels much below 2% in this cycle. Yields have traded in a tight range around an average 2% since September, including so far into 2012. We are now of the view that a break to the upside, to 2.25-2.50%, is likely and recommend going tactically short. Using Mar-12 futures contracts, which closed on Friday at 130-08, we would aim for a target of 126-00 and stops on a close above 132-00." We added the following: "As a reminder, don't do what Goldman says, do what it does, especially when one looks the firm's Top 6 trades for 2012, of which 5 are losing money, and 2 have been stopped out less than a month into the year." Sure enough, as we tabulated last night, those who had listened to this call, and also gone long stocks as Goldman urged on March 21, have lost nearly 30% in about 2 months. Those who listened to us and did the opposite, well, didn't. Which is why the just released note from the very same Garzarelli who 4 months ago was so gung ho on shorting bonds, just cut his bond yield forecast for the entire world, US Treasurys included: "We now see 10-year US Treasuries ending this year at 2.00% (from 2.50% previously, and 30bp above current forwards), rising to 2.50% (previously 3.25%, and 60bp above the forwards) by December 2013. The corresponding numbers for German Bunds are 1.75% and 2.25%." In other words, it is now that Doug Kass should have made his short bonds call: not when he did it, a month ago and got his face bathsalted right off. For those asking - yes: Goldman is now selling bonds to clients.
Here is the summary:
How does Goldman get there:
Reflecting an intensification of sovereign pressures in the Euro area, we are lowering our forecasts for the major government bond markets. We now see 10-year US Treasuries ending this year at 2.00% (from 2.50% previously, and 30bp above current forwards), rising to 2.50% (previously 3.25%, and 60bp above the forwards) by December 2013. The corresponding numbers for German Bunds are 1.75% and 2.25%.
A Recrudescence of EMU Risks
The month of May has seen a strong rally in interest rate markets in the developed economies, accompanied by a flattening in term structures. There have been two proximate drivers of the price action. The first is a modest decline in leading indicators of global industrial activity, as captured by our GLI. This has been reflected in both cyclical stocks and rates.
The second, and more dominant, force has been a resurgence of market pressures in the Euro area sovereign space, tied to developments in Spain and Greece. In the former, investors concerns relate to the recapitalisation of commercial banks, and the impact this could have on borrowing requirements. In the latter, a political stalemate has raised the spectre of a break away from the common currency union, reinforcing ongoing portfolio allocations within EMU towards bonds of the highest credit standing.
Germany has been leading other major rate markets: 30- year Bund yields have fallen more than 60bp since the start of May to below 2% (corresponding to a total return over this period of 15%). To put this in perspective, they are on par with 30-year yields in Japan—where domestic inflation is running 200bp lower. Even at these depressed yield levels, the ‘call skew’ on July Bund options for delivery on June 22, that is, after the second Greek parliamentary elections) has continued to increase.
Formal statistical tests confirm that, in a departure from historical norms, since March 2012 the causation has run from German yields to those in the US.
The flipside of such extreme shifts in German yields has been a meaningful increase in long-term spreads of Italy and, in particular, Spain relative to the core countries of EMU. Demand for intermediate maturity bonds from these two issuers continues to be low, and almost entirely domestic—as we had long expected. Intra-EMU government rate differentials now embody an element of currency risk. They are at levels far above what would be justified by relative economic fundamentals should concerns over the viability of EMU abate.
1. An Increasingly Bimodal World
When benchmarked against our central macroeconomic expectations, intermediate maturity yields in the main economies have reached levels that, on past norms, would lead us to recommend short positions.
Our Bond Sudoku framework, which provides us with an indication of how expected global macro factors should map into intermediate maturity government bond yields, suggests that German Bunds, US Treasuries and UK Gilts are now trading around 100bp too expensive relative to ‘fair value’, equivalent to a departure from the fitted value of well beyond one standard deviation.
We comment on the merits and limitations of Sudoku below, but the same conclusion is reached if we use alternative approaches tobond valuation.
The main issue we face at this juncture is that the distribution of possible macro scenarios, and hence that of expected government bond returns, is becoming progressively bimodal. Ahead of us, we think, lies a central macroeconomic scenario with quantifiable risks around it. Alongside this outcome, however, there is a much less probable but also much more harmful alternative, where depressed economic activity and uncertain institutional shifts in the Euro area interact. In this context, summary statistics such as mean and standard deviations could be deceptive.
2. Our Baseline Case: Gradual Progress Towards
Deeper Integration in the Euro area As we set out in a recent note (see The Euro area: 3 Greek Scenarios and Market Implications, May 28, 2012), our baseline case for Greece envisages: (i) a (painful) modification of the ‘troika’ program; (ii) continued ECB funding for local banks under Emergency Liquidity Assistance (ELA); and (iii) the rest of Euro area slowly and discontinuously progressing towards deeper integration in the areas of financial and banking regulation and fiscal policy.
Such integration encompasses all the main ‘desiderata’ we have written about in previous research. These include the EFSF helping Spain out under a Memorandum of Understanding specific to bank recapitalisations (i.e., not a full macroeconomic adjustment program); a common Euro area wide bank resolution authority regime and, eventually, a merger of deposit insurance schemes; a political agreement to reach a mutualisation of legacy debt subject to conditionality and collateral. These are admittedly ‘end game’ solutions, which can be implemented over the space of several quarters, rather than weeks. But markets would be reassured by theanchoring of medium-term expectations around a policy objective shared by all member countries.
In this world, the very large insurance premium priced into German Bunds and US Treasuries since last summer should gradually dissipate as Europe slowly becomes more integrated and resilient to events in Greece, allowing bond yields to realign themselves to their macro underpinnings.
Within this central case we would expect peripheral spread curves to remain relatively steep, underpinned by the ECB’s ‘term liquidity on demand’ policy. Both the level and slope of intra-EMU spreads would interact primarily with shifts in the growth outlook. An exercise of mapping spreads to fundamentals predicated on the baseline scenario would see 10-year differentials with Germany head towards 250-300bp for Italy (currently 465bp) and 300-350bp for Spain (currently 530bp) on a 6- to 12-month horizon.
It is conceivable that more positive developments could take place. We have, for example, discussed the merits of the European Debt Redemption Fund, which we outline again in the Focus section. Any hint of the Euro area moving in these directions would result in a sharper rise in German yields towards their macro equilibrium, and a compression of German swap spreads to 20bp-25bp from 50bp currently), with 10-year Italian bonds trading around 150bp-200bp over mid-swaps.
3. And a Darker Mode
But there is also an alternative scenario, in which a withdrawal or expulsion of Greece from the single currency could result in a hit to the Euro area’s aggregate GDP of around 2%, assuming robust policy responses. Failing those, the risk of a broader collapse of EMU could be material, and the contraction in economic activity approach double digits. It is this sort of ‘rare event’ risk that has led to a decline in traded volumes over the past months, a build-up of cash balances and a ‘flight-tosafety’.
Arguably, high credit quality government bonds incorporate a greater chance of bad outcomes occurring than other assets. The 3-month rate in 10-years’ time in Germany is now priced at 2.0%, at the ECB’s area-wide inflation rate target. The corresponding number in the US is 2.3%, approaching levels seen after the collapse of Lehman Brothers, and in the UK is 3.0%.
Nevertheless, in a world where the demand for creditriskless fixed income assets outstrips supply (the latest IMF Financial Stability Report provides a comprehensive overview), high rated government securities could be in greater favour than we have so far assumed.
Which all leads to...
Our Forecast Changes Reflect a More Gradual Adjustment
Drawing on these observations, we are lowering our major markets’ interest rate forecasts to reflect a more gradual adjustment of intermediate nominal yields towards values consistent with their macro underpinnings.
- We now see nominal 10-year US Treasuries ending this year at 2.00% (from 2.50% previously) and then gradually rising to 2.50% by December 2013 (previously 3.25%), approaching our Sudoku measure of ‘fair value’ from below. At the time of writing, with spot yields at 1.6%, our end-2012 forecasts are 30p above the forwards, and those for end-2013 are 60bp above the forwards.
- The corresponding numbers for German 10-year Bund yields are 1.75% in December 2012 (previously 2.25%), 40bp above the forwards; and 2.25% (previously 3.00%) in December 2013, 70bp above the forwards.
- We forecast 10-year UK Gilts at 2.00% in December 2012 (previously 2.75%), 25bp above the forwards; and 2.75% (previously 3.50%) in December 2013, 75bp above the forwards.
- Lastly, we see JGBs at 1.00% in December 2012 (previously 1.25%), broadly in line with the forwards, and 1.30% in December 2013 (previously 1.70%), or 20bp above the forwards.
- Along similar lines, we are also lowering the forecast for 10-year rates on the bonds of Canada, Australia, Switzerland and Sweden, as summarised in the Table below.
- We see two main reasons why bond yields should increase over the forecast horizon:
- First and foremost, we expect a relaxation of strains in the Euro area, consistent with the baseline case sketched out above. As already mentioned, however, it may take more time for flight-to-safety flows to reverse given the damage to investors’ confidence experienced over the past year.
- Second, we are forecasting a sequential improvement in nominal GDP growth in the advanced economies. Our present forecasts envisage a quarterly annualised expansion from 2.75%-3.00% in the first half of this year to 4.75%-5.00% between 2012Q4 and end-2013. Although we do not expect any of the major central banks to adjust policy rates upwards over this horizon, markets may start discounting changes over 2014-16.
And so on.
Bottom line: take whatever Goldman say, and flip it.