QE3 ON: Goldman Lowers Global Government Bond Forecasts Following 2012 US GDP Cut To 2.1%, Repeats "QE3 Is Part Of Baseline Estimates"

Tyler Durden's picture

For those wondering why gold just surged by about $20 dollars, and why Gartman's cab driver once again proves to be far more astute than his passenger, we bring to your attention  a report just released by Goldman's Francesco Garzarelli which is appropriately titled "The Price of Slower Growth" - appropriately, because in it Goldman slashes the firm's outlook on global policy rates across the board, slashes to cut its 10 Year bond yield outlook from 3.75% to 2.75% in 2011 and from 4.25% to 3.50%, slashes 2012 US GDP from 3.0% to 2.10%, and once again makes it all too clear that QE3 is coming, and not only coming but is already priced in (to the tune of about $300-400 billion): "In previous work, we have estimated that every US$1trn in purchases, if maintained, decreases 10-yr Treasury yields by 25bp-50bp. If our subjective assessment that market participants now assign a greater-than-even chance of ‘QE3’ is correct, and considering that the expected ‘unsterilized’ size of these purchases is in the region of US$600-800bn, this would equate to as much as 20bp being already ‘in the price’. Clearly, these magnitudes are unobservable, and thus subject to great uncertainty. Nevertheless, our calculations would suggest that the bond market is already discounting a mild recession and the chance of a Fed reaction to it." Translation (and this is nothing new to ZH readers): Bill Dudley has his marching orders from Jan Hatzius: GS now sees deflation as the broader risk, and anything and everything must be done to make sure Wall Street has another record bonus season round, pardon, deflation must be halted.

Updated Goldman GDP forecast table:

And where GS sees 10 Year rates:

Full note from Goldman Sachs:

The Price Of Slower Growth

On the back of the downgrade to our growth forecasts, published on August 5, we have revised lower our 10-yr bond yield projections across the board. Nevertheless, we find that intermediate yields are very depressed, particularly in the US, even taking into account our new macro projections. This indicates that there is now a substantial ‘tail risk’ discounted in high-quality sovereign bonds, alongside expectations of further asset purchases by the Fed. Developments in Euro-zone bond markets have been momentous over the past month. In line with our expectations, bond prices in Greece, Ireland and Portugal have increased from distressed levels following a substantial modification to official-sector loans. Italian and Spanish bond yields remain elevated relative to Germany’s and France’s. The introduction of domestic fiscal and structural policies, important steps towards unconditional ‘risk-sharing’ among EMU members (agreed on July 21), and open market interventions by the ECB should all help underpin the attractive valuations of both BTPs and Bonos.

A Precipitous Rally

After a relatively lacklustre first half of the year, returns on intermediate maturity government bonds in the advanced economies have risen since the start of the third quarter, and particularly going into the month of August. The ‘winners’ have been those markets most sensitive to cyclical swings (7-10-yr US Treasuries have returned almost 7% non-annualized since the end of June, with yields falling 50bp since the start of this month; returns in the UK and Australia have been commensurately high). By contrast, returns on traditional ‘safe havens’ have been mixed: Swiss bonds have risen strongly, but Japan’s have moved sideways in local currency terms. The performance of bonds issued by EMU states has diverged significantly, although this was partly reversed by the ECB’s resumption of direct purchases.

An unusual combination of interconnected economic and political factors has driven the rally, some of which are exceptional in nature:

First and foremost, these moves have come against a backdrop of a substantial decline in growth expectations, following a disappointing set of data, particularly in the US. To be fair, the growth data have been less uniformly bad over the past two weeks, as captured by our US Current Activity Indicator. Thus, the recent sharp price action in equities and bonds has probably more to do with a focus on ‘tail risks’ in an anaemic US growth environment, concerns about a broader global growth slowdown (as reflected in the sharp moves in global indices and cyclical currencies) and positioning considerations, rather than outright weakness in the incoming data.


Second, the macro backdrop and the tightening in US financial conditions that it caused precipitated a further round of policy easing by the Fed. The central bank has indicated that its current economic forecasts warrant policy rates remaining close to zero for at least another two years and that it stands ready to expand its balance sheet further if needed. We have built a third round of long-term asset purchases (‘QE3’) into our baseline (see US Daily: For More Easing, Will Fed Go Big or Go Long?, August 15, 2011), although that is, of course, contingent on sub-trend growth in the near term. Judging from our client interactions, and recent investor surveys (e.g., CNBC/Liesman), this is now seen as the most likely scenario, although it is far from being a universally held view.


Third, ‘safe-haven’ flows originating from tensions in Euro-zone sovereign markets have increased in the wake of a restructuring of Greek sovereign debt, spreading to Spain and Italy, which jointly represent around one-third of the area’s aggregate government bond market. Unsecured bank funding markets in the Euro-zone have seized up even for French issuers.


Lastly, the downgrade of the US sovereign debt from AAA to AA+ (negative outlook) by S&P has added to the anxiety surrounding sovereign and financial risk globally, and supported the view that a further relaxation of fiscal policy to support demand will be politically harder to achieve.

In conjunction with the downward revisions to the outlook for growth and policy rates in a number of countries, as detailed in Table 1 above, we have revised down our government bond forecasts across the board. Most notably, we now see 10-yr US Treasuries ending this year at 2.75%, from 3.5% previously (see Table 2). Yet, our bond return expectations remain moderately negative, particularly beyond the next three months. This is for two main reasons:

First, our measure of the global bond risk premium is now at its lowest level since the start of the credit crisis in 2007—(this is the ‘common factor’ jointly driving yields of the top-rated government bond markets after controlling for country-specific contemporaneous macroeconomic conditions such as industrial production, CPI inflation and short rates (see Chart 1).

Admittedly, many investors and businesses are considering contingency plans in light of recession risks, and are understandably willing to pay for owning counter-cyclical assets. Furthermore, ongoing strains in the Euro area have led to reallocations into German bonds. That said, the major central banks continue to provide large amounts of liquidity and collateralized funding. And there is a progressively deeper ‘sharing’ of sovereign and financial risk across the Euro-zone countries (e.g., the large conditional transfers to the program countries and a greater latitude for the EFSF to intervene in support of sovereigns and banks of the Euro area, as agreed at the Euro-zone Summit on July 21). Against this backdrop, highly rated government securities appear already to assign a significant weight to ‘tail outcomes’.

Second, a meaningful hit to future activity growth has now been incorporated into cyclical assets, according to our metrics, particularly in the US. Consider the following:

Real rates are negative across the term structure. US 5-yr nominal rates deflated by actual core inflation are below zero (and deeply negative when deflated by expected inflation), and on this measure have been below Japan’s for several months. At around 3.70%, nominal 5-yr/5-yr forward rates are below the rate of nominal growth at that horizon (which consensus puts above 4.50%).


Treasury yields are already discounting a very weak outlook... Using our revised global macro projections as inputs, our Bond Sudoku model indicates that 10-yr US Treasuries should currently trade at 3.0% (see the Table on page 12). The ‘fair value’ at the end of this year rises to 3.2%, or around 60bp above the forwards. Using an alternative recession scenario in the US and Euroland, we generate fitted yields of around 2.2%-2.5%, increasing to 2.6%-2.8% by year-end, still above the forwards.


...and/or a high chance of further asset purchases. As mentioned earlier, the Fed has indicated that it stands ready to review the ‘size and composition’ of its securities holding (currently amounting to around US$2.6trn, or US$1.6trn above the ‘normal’ size). Given that short rates are now more strongly underpinned by the Fed’s conditional commitment to keep policy rates close to zero until at least mid-2013, the impact of purchases would be most evident in intermediate and longer-dated securities. In a ‘portfolio balance’ approach (see the April issue of this publication for a discussion), the impact on yields of ‘unsterilized’ purchases of Treasuries (an increase in the size of the balance sheet, involving a corresponding rise in the money supply) or ‘sterilized’ ones (via the sale of shorter-dated instruments, resulting in a change in the composition of the asset portfolio towards longer-duration assets) should be close substitutes. In previous work, we have estimated that every US$1trn in purchases, if maintained, decreases 10-yr Treasury yields by 25bp-50bp. If our subjective assessment that market participants now assign a greater-than-even chance of ‘QE3’ is correct, and considering that the expected ‘unsterilized’ size of these purchases is in the region of US$600-800bn, this would equate to as much as 20bp being already ‘in the price’. Clearly, these magnitudes are unobservable, and thus subject to great uncertainty. Nevertheless, our calculations would suggest that the bond market is already discounting a mild recession and the chance of a Fed reaction to it.

Based on these considerations, we do not see scope for 10-yr US bond yields to break below 2.00%-2.25% in this cycle. Conditional on our baseline forecasts, we expect them to rise to 2.75% by year-end and further above the forwards in 2012. Intermediate rates in other major countries should also increase, although the starting level is somewhat less stretched.
Sovereign Tensions in the Euro-zone

Developments in Euro-zone bonds have been momentous since the start of July. Official-sector loans to Greece, Ireland and Portugal (ranked pari passu with outstanding securities) have been substantially modified, implying a large wealth transfer from the Euro-zone’s taxpayers to existing holders of government bonds and resulting in a significant decline in yields (see Chart 4). In return, the Euro-area has managed to limit the ‘collateral damage’ from a sovereign default, and has increased its control over public-sector cash flows in these three member countries. This broadly fits into our broad theme of a ‘managed deleveraging’.

However, the involvement of the private sector in a second rescue package to Greece (through debt exchanges, maturity extensions or buybacks) has set in motion an unintended chain of events. Credit rating agencies (CRAs) have suggested that the same treatment will be applied to Ireland and Portugal, downgrading both these issuers. Moreover, CRAs have called into question the ability of Italy and Spain—systemic borrowers with large roll-over needs—to access markets continuously, leading to a sharp and self-reinforcing increase in their yields.

France has also come under the spotlight recently, and its bonds and CDS have largely underperformed Germany’s (the 5-yr cost of protection is now at around 150bp, comparable to Brazil’s). In sharp contrast to Italy and Spain (where the covariance between sovereign and financial risk has been positive), France’s government funding costs have declined as pressures on domestic banks intensified.

There have been three connected policy responses to these developments.

On the domestic fiscal front, Italy and Spain have announced additional fiscal tightening measures, alongside structural reforms including a ‘German-style’ balance-budget rule in the Constitution and changes to labour market negotiations. Although the short-term impact on domestic demand is likely to be negative, trend growth should instead be positively affected. France is also due to announce further fiscal measures by the end of this month.

At the Euro-area level, the EFSF has been given authority to intervene in secondary bond markets and in the recapitalization of banks outside ‘program countries’. These measures imply a higher degree of ‘risk sharing’ among EMU members, which could potentially represent a key step towards greater centralization of at least some fiscal functions currently conducted by the individual states.

Finally, upon reaching guarantees that the abovementioned fiscal policies will be carried out speedily, the ECB has extended its Securities Market Program to Italy and Spain, purchasing government bonds of these two sovereigns in the open markets. Around EUR25-30bn are estimated to have been purchased so far. The resulting injections of liquidity are sterilized, although this is not a binding constraint in the context of the central bank’s ‘full-allotment’ refinancing policy. We estimate that purchases could cumulatively amount to EUR100-130bn, roughly equating to the net borrowing needs of these two sovereigns for this year.

All these policies carry implementation risks, adding to volatility in the markets. The Greek debt exchange is planned for the beginning of September and later that month the measures agreed at the Euro-zone Summit of July 21 will be brought before national Parliaments for approval. The level of political support for both domestic fiscal policies and those at the Euro-zone level has been at times discontinuous.

On balance, our view is that the substantial deviation from (improving) fundamentals, with the ECB’s open market purchases acting as a backstop, should allow Italian BTPs and Spanish Bonos to converge slowly back to their historical relationship with macro factors (see Charts 5-7). We see the 10-yr BTP-Bunds spread trading in the region of 225bp-250bp into year-end, and the corresponding differential in Spain to be 200bp-225bp. We also view pressures on France as overblown. A further tightening of sovereign spreads should result from more tangible progress towards greater coordination of fiscal policies, providing credible counter-cyclical demand management in EMU member countries.