9 August 2010

Economics
Beyond the Cycle
www.sgresearch.com

American Themes
QE 2, a fire drill and preparations for the real thing
Stephen Gallagher Chief Americas Economist (1) 212 278 4496
stephen.gallagher@sgcib.com

Fed Chairman Bernanke in testimony before Congress in late July stated that the Fed would take additional steps to support the economy as needed. In follow-up he offered further steps that may be helpful but difficult to describe as aggressive. His FOMC colleague, St Louis Federal Reserve President James Bullard has argued in favor of more aggressive action, additional Quantitative Easing (QE), if the risks of deflation intensify.

Aneta Markowska US Senior Economist (1) 212 278 6653

aneta.markowska@sgcib.com

Martin Rose Research Associate (1) 212 278 7043

martin.rose@sgcib.com

When Bullard proposed a more aggressive second round of quantitative easing the first questions were: what would the Fed purchase and how would that
Conventional Wisdom

influence the market? Our immediate response - and it seems the markets response - is to believe the Fed would purchase Treasury securities along the yield curve, rather than purchase
Fidelio Tata Head of US Rate Strategy (1) 212 278 6213
fidelio.tata@sgcib.com

additional mortgage-backed securities. The Treasury curve would be mostly biased to flatten, but there would be limited interest in the long bond. It is striking that prior to any announced new QE program, the market already seems to be trending along these conventional lines. Additionally, the liquidity offered would support most assets and likely weaken the dollar. That market is trading on hopes for more QE. There may be danger in this conventional wisdom. An announced QE program may continue these trends further, at least initially. What are the pitfalls of conventional wisdom? Too often there are many.

Implicit assumptions behind initial market performance and what may be wrong.
1. The biggest and perhaps the most dangerous assumption is that QE will not have much success initially in altering inflation expectations. Suppose that investors viewed the Fed s new QE program favorably. Expectations for the economy could rally. Bond yields could go up. Eventually a successful Fed reflation effort could drive yields notably higher with the question being, when will inflation expectations turn more constructive? Table of Contents: What would the Fed want to accomplish with a new QE round? The important Page 2 role of 2. The second assumption is that short-term yields near zero cannot go much lower and that additional Fed purchases would drive all yields toward zero and therefore flatten the curve. This thought is recognizing the introduction of a new Treasury buyer with deep pockets and long staying power. Does it miss, however, the opportunity for current owners of Treasury securities to unload when a large buyer is active? A few funds, and foreign investors, could be looking to sell into any purchase program, particularly if the QE program is viewed as possibly succeeding.

inflation

expectations Page 3

Executing further QE if needed The Fat-Tail Page 4 Scenarios Page 8

The announcement and initial rounds of Treasury purchases by the Fed would lower yields and flatten the Treasury curve. This is more due to a zero-bound constraint and the already distant market expectation for rate hikes by the Fed. This scenario already priced in the market - implies that inflation expectations are stable or falling. The high risk here is that once the market perceives potential success from QE for the economy, inflation expectations could rebound swiftly.

Macro

Commodities

Forex

Rates

Equity

Credit

Derivatives

Please see important disclaimer and disclosures at the end of the document

This document is being provided for the exclusive use of MARTIN ROSE (SGCIB)

What would the Fed want to accomplish with a new QE round?

The end goals are easy. Faster growth, lower unemployment and a stable inflation rate between 1.5% and 2.0%. Reaching those goals via QE is not as straightforward. We see three desires the Fed might hope to achieve: lower rates, raise inflation expectations or alter the slope of the curve. 1. Lowering rates. The Fed can have significant influence on rates for any instrument it chooses to buy. The Federal Reserve is most likely to purchase Treasury securities (see below for further comment). The Fed can simply buy large quantities to drive the price or yield to a desired level. This may not help the economy. Market rates on corporate and mortgage debt would in time widen to the Treasury curve the further Treasury yields moved from fair value. Also, non-Fed holders of Treasuries would over time sell over-valued securities. 2. Raising inflation expectations. Inflation expectations are a highly significant variable determining outright inflation. Deflation episodes require expectations of falling prices. Anchored and positive expectations are the most desired outcome. Raising inflation expectations could raise market interest rates. Market rates are comprised of real interest rates plus inflation expectations. Real rates could fall. They seem to be near zero already, but real interest rates can and have fallen below zero. The Fed can lower interest rates across the curve if it buys sufficient amounts of Treasury securities. If inflation expectations rise, the Fed may find itself the only buyer of Treasury securities and nearly all other interest rates would widen to the Treasury curve. These are extremes, but need to be thought out. Ultimately, monetization becomes a worry that raises serious inflation concerns. In Japan, monetization has not led to inflation. We will be examining this issue more fully in later writings. 3. Altering the slope of the curve. The Feds Open Market Desk does not on its own volition take actions to alter the curve. That would be a policy choice. The FOMC could charge the Feds Open Market Desk to embark on such an approach, but this has been very rare. We show below how the Fed has purchased securities in the past in a manner intended to be curve neutral. A heavy buying program of Treasury securities when the short-end is already near zero, however, could readily impact the curve. The Feds key motivation is to fight deflation and its fear is that inflation expectations could slip and even turn negative. That would be the recipe for serious deflation. Hence raising inflation expectations or at least preventing a further drop of inflation expectations is how we see the Fed progressing. So far, and initially, inflation expectations are low and could fall before rising. New policy stimulus, particularly QE, would have more influence on the supply/demand for Treasuries (removing supply available to the market) without much change on inflation expectations. This may not hold long. Signs of success in the Feds efforts would eventually raise inflation expectations. We saw this transpire in the market in mid-2009 as the economy began to strengthen. Governments have not seen much change in their borrowing costs, although municipal yields failed to follow Treasury yields down over the past month.

9 August 2010

This document is being provided for the exclusive use of MARTIN ROSE (SGCIB)

American Themes

The important role of inflation expectations

The self-fulfilling nature of price expectations is a serious threat for both high inflation and deflation. 10-year Treasury yields were well below 3% in the US at the end of last week, following economic news as well as Bullards comments about QE. See mingly, this is due to falling inflation expectations. Bullard tied the increasing risks for deflation to low inflation expectations. What are current inflation expectations? There are different measures, but generally all are sliding. Our models for inflation depend heavily on inflation expectations (See American Themes, Labor market slack and the growing deflation risk, March 2009). Deflation, a sustained price decline, needs falling expectations.

Inflation Expectations Measures

Consumer Survey (UM)
6 % 5 4 3 2 1 0 02 03 04 05 06 07 08 09 10 Inflation Expectations 5-10yrs Inflation Expectations 1yr

Current measures show softening inflation expectations

There are three sources of information about inflation expectations: surveys of consumers, surveys of inflation forecasters and the market for Inflation-indexed bonds. Economists inside and outside the Fed look at all three measures. Rather than precision on the number, policy making is formed via direction or movement of expectations. 1. Consumer surveys. The most popular is the University of Michigan survey that offers both short-term and long-term expectations. The consumer survey can be susceptible to gasoline prices. On the good side, these expectations have been surprisingly accurate.
1-yr Survey Prof Forecasters 10-yr SPF

Professional Forecasters Philadelphia Fed

3.0 % 2.5 2.0 1.5 1.0 0.5 0.0 02 03 04 05 06 07 08 09 10

2.

Professional forecasters. There are many surveys. When the Fed sources private forecasters it cites the blue-chip economic forecasts. Surveys done by the Wall Street Journal or Bloomberg can be easy substitutes. In our models of long-term inflation that incorporate inflation expectations, we use the Philadelphia Federal Reserve Survey of Professional Forecasters because of its long-term compilation.

TIPS Break Even and Fed Adjusted 5yr-5yr forward

4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 05 06 07 08 09 10 Implied by TIPS Implied by inflation swaps %

3.

TIPS market. Economists and Fed officials know the data is flawed by liquidity. The 10-year TIPS benefits from a liquidity premium relative to other TIPS issues. The Fed publishes an adjusted TIPS inflation forecast that can be found on Bloomberg and charted on the left. Brian Sack, head of the Feds Open Market Desk in New York, was one Fed official behind the method to adjust for liquidity. Fed analysts may have their own methods for interpreting the basic break-even inflation rate (BEIR) in the TIPS market, but all generally know the basic BEIR. It is movements rather than precise levels that generate concern. The basic BEIR is slowing more than the inflation adjusted. Either the premium for 10-year TIPS is improving relative to other TIPS issues or the adjustment is failing to capture the drop of inflation expectations. The Fed and private analysts need to look at both without overweighting one.

Sources: Global Insight, Bloomberg and SG Cross-Asset Research

Overall inflation expectations are softening. Short-term expectations are seeing a more pronounced decline since energy prices have recently softened. Energy prices and other commodity price are enjoying a renewed upswing. Rising oil prices again above $80/bbl should lift headline data. Long-term inflation expectations have been more anchored.

9 August 2010

This document is being provided for the exclusive use of MARTIN ROSE (SGCIB)

What moves expectations of inflation?

In academia, we consider rational or adaptive expectations. The reality is more mysterious and the stability of long-term inflation has been as much a mystery as swings or movements of inflation expectations. Central Bankers point to CB credibility as a factor behind low and stable long-term inflation expectations. That may be partly true. Not all Central Banks enjoy the same credibility and the inflation outlook has been low and stable on a global basis. The dollar, commodity prices, actual price (CPI) performance and excess capacity in the system are top factors that influence inflation expectations, at least over short-term horizons. The recent lift in some commodity prices, particularly foodstuffs, may push consumer expectations higher. Also, housing costs, a key component of the CPI, may be bottoming, and if so could have a very significant influence on expectations. Fed officials, as suggested by the concern voiced by St Louis Fed President Bullard, are more worried about the downward trend of inflation expectations. They are watching the broad array of inflation expectation measures charted above and will be increasingly likely to offer further quantitative easing if these measures decline.

Executing further QE if needed

Quantitative Easing, in a very generic definition, is growing the Fed balance sheet via the purchase of some assets. The Federal Reserve has executed QE via purchases of Agency, Agency-MBS and US Treasury debt. The heavy concentration of earlier QE efforts was in the Agency-MBS market where the Fed purchased $1.25 trillion of Agency-MBS. From November 2008 to March 2010, the Federal Reserve purchased $1.25 trillion Agency MBS, $175 billion Agency debt securities, and $300 billion Treasury securities.
Fannie/Freddie yields spreads to 10-year swap
250 bps 200 150 100 50 0 05 06 07 08 09 10
Source: Bloomberg, SG Cross Asset Research

US mortgage yields 30-year fixed rate

8.0 % 7.5 7.0 6.5 6.0 5.5 5.0 4.5 03 04 05 06 07 08 09 10 30yr Agency conforming FRM 30yr jumbo FRM

Fed announces it will buy Agency and Agencybacked MBS.

Fed boosts amount of Agy MBS debt it will purchase

If further rounds are undertaken, what type of securities and how much would they purchase? Treasury purchases are far more likely than additional Agency purchases. Agencies are predominantly Freddie-Mac and Fannie-Mae backed securities. When the Fed first began its Agency debt purchases the mortgage market was broken and the absence of buyers for Freddie-Mac and Fannie-Mac debt threatened the entire mortgage origination market. Spreads on Freddie and Fannie debt were exceptionally wide and mortgage rates for potential home buyers were also very wide to the Treasury curve. The restoration of flows to Fannie and

9 August 2010

This document is being provided for the exclusive use of MARTIN ROSE (SGCIB)

American Themes

Freddie and the normalization of mortgage rates were among the most successful undertakings by the Fed to support financial markets and the economy during the crisis. Mortgage rates are now down significantly and have normalized more or less to the Treasury curve. The Treasury is making large loans available to the Agencies in addition to market funding. Going forward, the Fed is more likely to buy the large quantity of debt that the Treasury needs to issue into the marketplace.

Recent experience of Treasury purchases by the Fed

The Treasury portion of the Feds QE program amounted to $300 billion from March 2009 to November 2009. The size of the operation was not very aggressive on two points. First, the Treasury has been issuing more than $1.4 trillion net new Treasury securities each year for the past two years. Second, the Treasury held $700 billion of US Treasury securities at the start of the crisis in August 2007 and allowed $300 billion to roll off when it made massive loans to the banks. The $300 billion in new purchases only raised the Feds Treasury holdings back to their pre-crisis levels. Nonetheless, the purchases are very important for the market as they occur.
Fed holdings of Treasuries ($bn) Fed holdings of Treasuries (% of total)

Treasuries held in SOMA Portfolio ($ bn)

less than 2 years

Treasuries held in SOMA Portfolio (%)

less than 2 years

47
51 46 211

2-5 years 5-7 years 7-10 years

6% 7% 6% 27%
2-5 years 5-7 years 7-10 years

97
10-15 years
15-20 years

12%
10-15 years 15-20 years

118

207

20-30 years

15%

27%

20-30 years

Source: Federal Reserve, SG Cross Asset Research

Key points on Treasury purchases by the Fed:

1. The Fed asks dealers for offers of Treasuries in specific maturity ranges for each open market transaction. In old jargon this was called a bill-pass or a coupon pass. After dealers submit offers, the Fed announces the specific issues, amounts and prices it will take. 3. The Fed spreads its purchases out over the curve. Historically the Fed owned a piece of each Treasury offering. The exception has been the crisis when private investor demand for Treasuries was very strong and the Fed allowed its bill holdings in particular to roll off. 4. The Fed open-market desk does not try to influence the shape of the curve. It would only do so if directed to by the FOMC. In the past, Operation Twist was a cooperative effort between the Fed and Treasury to re-shape the curve. The unusual environment could revive such a directive of the FOMC, but unlikely. Even if not intended, massive Treasury purchases would likely have unintended consequences.

2.

9 August 2010

This document is being provided for the exclusive use of MARTIN ROSE (SGCIB)

From March 2009 to November 2009 when the Treasury purchased $300 billion of Treasury securities it purchased across the yield curve. The Treasury purchased greater amounts at shorter maturity levels and generally aimed to duration-weight the purchases to neutralize their impact on the curve. The charts below show recent Fed purchases of Treasuries by maturity. The chart on the left shows the notional amount of Treasuries held in various maturity buckets of the SOMA portfolio. It appears that the portfolio is front-loaded with comparably larger positions in shorter-dated Treasury securities. However, the duration of those short-dated securities is smaller than the duration of longer-dated securities. Thus, the curve impact is different across the curve for purchases of equal amounts. To normalize the curve impact we multiply the position values by their respective DV01 (Dollar value of a basis point change in yield). The result is shown in the chart on the right. It is quite obvious that the SOMA portfolio is very well balanced with respect to the curve impact on different parts of the yield curve. A similar DV01 for each maturity bucket implies that the curve impact of the purchase (and a possible liquidation in the future) is approximately the same across maturities. Put differently, the SOMA purchases were weighted to cause parallel yield curve movements. They are basically curve-impact-neutral.
2009 Fed purchases of Treasury securities.
USD bn 250 200

Dollar value of a basis point change

USD 90,000 80,000 70,000

Remaining Maturity of Treasuries held in SOMA Portfolio

DV01 of Treasuries held in SOMA Portfolio

150
100 50

60,000 50,000 40,000 30,000 20,000 10,000

less than 2 years 2-5 years 5-7 years 7-10 years 10-15 years 15-20 years 20-30 years

less than 2 years

2-5 years

5-7 years

7-10 years 10-15 years 15-20 years 20-30 years

Source: Federal Reserve, SG Cross Asset Research

The Treasury data is charted by looking at their remaining maturity and not the original issue maturity

Influence on the yield curve

The Feds intent may be neutral but the purchases can have a strong influence on the curve nonetheless. Markets constantly re-assess the economic forces and Fed rate projections. Initial purchases may coincide with the market pushing out further to the future its expectations that the Fed will raise rates. Additionally, these actions may increase confidence that the economy will rebound. Inflation expectations might rise. These factors would steepen the yield curve. The expectation for a flatter curve - at least immediately on the prospects for further QE - is based on a belief that shorter-rates are near zero and are zero-bound. At an extreme, purchases that drive the entire curve toward zero would flatten the curve at that one level. This thinking may be dangerously simplistic. At some point, the market should gain confidence that the Feds efforts to restore growth and support inflation expectations are succeeding. At that critical juncture, the Treasury yield curve should steepen. Historical information is thin, but the recent Federal Reserve and the Bank of England purchases offer some insight.

9 August 2010

This document is being provided for the exclusive use of MARTIN ROSE (SGCIB)

American Themes

Government yield curve moves following CB purchases

We examine in the charts below the slope of the Treasury and the Gilt curve following purchases of government debt by the Federal Reserve and the Bank of England. There are similar steps
Treasury yield curve slopes during Feds QE program
US Treasury Spreads
+ 50 bn Announcement: 75 bn

Gilt curve slopes during BOEs QE program

UK Gilt Spreads
+ 50 bn + 25 bn Program completed: 200 bn
1/10 3/10 250 bps

Announcement

Program completed

bps 200

200

150

150

100
100

50 2Y-5Y 0 1/09 3/09 5/09 7/09 9/09 11/09 1/10 3/10

Source: Federal Reserve, Bank of England, Bloomberg and SG Cross Asset Research

50

5Y-10Y

10Y-30Y
0 1/09 3/09

2Y-5Y 5/09

5Y-10Y 7/09

10Y-30Y 9/09 11/09

In both the Fed and the BoE examples, the commencement of QE steepened the curve, particularly in the short-to-mid coupon sector. This may have coincided with a major readjustment of Fed and BoE rate expectations for the future. The announcements of these programs should have given the markets reason to believe central banks would keep rates lower for a much longer period of time. In contrast, for the long-coupon spread here measured from 10-yr to 30-year in both cases the announcement and purchases induced some narrowing or flattening at the very long-end of the curve. The narrowing spread between the 10-year and 30-year is less pronounced in the US than in the UK. For the US, after some initial steepening in the first few months of the purchase program, the curve stabilizes. At the end of the program, as the economy was gaining traction in late 2009 and early 2010, the curve steepened, particularly in the 5-10 year sector which may have leaned more on rising inflation expectations and projections that for the near term the Fed would remain on hold.

Curve for the future in another round of QE

The market has already priced in an extended period of unchanged policy by the Federal Reserve. The 2-year note is already at all time lows near 50 bps. Another 25 bp drop could take place if the Fed signals a more prolonged period of low rates. The 10-year and even the 30-year have scope for larger absolute yield declines. It is mostly for this reason we expect the curve to flatten further. This would be true for the 5-10 year sector and the 10-30 year sector of the curve. These may be the initial moves. As the program completed and investors turned more confident on the economic outlook, the 10-year to 30-year spread in both the US and the UK steepened.

9 August 2010

This document is being provided for the exclusive use of MARTIN ROSE (SGCIB)

Recent activity
US Treasury yield curve more recently
400 Basis points 350 300 250 200 150 100 50 0 Jan-09 10-yr 2 yr 30 yr -10 yr 30 yr -2yr

Apr-09

Jul-09

Oct-09

Jan-10

Apr-10

Jul-10

Oct-10

Source: Bloomberg, SG Cross Asset Research

More recently, as the economy began to show signs of losing momentum and as speculation on further QE from the Fed gained serious traction, the 2-year to 10-year sector of the Treasury curve flattened. This might be due to the zero-bound constraint on the 2-year. Although still above 50bps, the zero-bound may be increasingly important. The 30-year yield however is widening to the 10-year. This might be explained by 1) higher very long-term inflation expectations, 2) the market does not expect aggressive Fed purchases in the very long end of the curve, or 3), the 10-year sector is benefiting uniquely, perhaps for liquidity reasons. In regards to the Fed purchase sizes, we have heard various officials express concern about holdings of long-term securities and the difficulties they presented for exit strategies. The reinforcing message, however, of our current analysis is that the Fed will remain duration neutral across the yield curve.

The fat-tail scenarios

A. ENDGAME HYPOTHESIS The automobile and banking industries have survived the meltdown and housing will eventually stabilize. This could be the last push of the "hedge fund Fed" to load up on Treasuries and mortgages. With less liquidity among the other market participants (hedge funds side-lined, bank prop-desks' wings clipped by regulation, Asia in wait-and-see mode etc.), this creates a historic opportunity for holders of TSY securities to cash out. Thus, the Fed buying could well be immunized by an equal amount of (motivated) selling from money managers, foreigners etc. B. YIELD CURVE HYPOTHESIS An inverted yield curve does not make the Fed look

good. Inverted yield curves have been associated with the recessions of 2000, 1991 and 1981. The curve also inverted in 2006. Because the Fed needs to hike front-end yields eventually, the Fed probably prefers a steep yield curve so that the curve does not invert right away (and signals another recession). Also, a steep curve creates income for financial institutions, which became junkies to the easy-money environment of the Fed and may react adversely if the curve inverts. Thus, the Fed will likely try to

9 August 2010

This document is being provided for the exclusive use of MARTIN ROSE (SGCIB)

American Themes

avoid a yield-curve-flattening. We remarked earlier that the Fed is neutral on the curve unless directed by the FOMC. The watch here would be that the FOMC issues an historic directive. C. CORPORATE ISSUANCE HYPOTHESIS August 2nd was the third-largest day of corporate issuance in 2010. IBM issued a 3-year with a 1% coupon, which is the lowest coupon for corporate deals ever. Take-away: corporate issuance makes sense from the issuers' perspective. This creates a huge liquidity pool for investors willing to switch from TSY into corporate credit. Liquidity/supply attracts buyers (eventually) and there could be a massive switch from TSY into credit on the horizon. The swap curve trades through the TSY curve for 10+ years, so there is plenty of room for corporate securities to cheapen (while issuance picks up) while remaining attractive from an issuer's perspective. D. TREASURY RISK HYPOTHESIS Treasury instruments used to be considered the closest proxy for being risk free. A number of recent developments question this assumption. First, swap spreads turned negative, first for 30-year and then for 10-year. One explanation is that holding a Treasury (and financing it in repo overnight) is now considered more risky than receiving on a long-dated swap. In a swap there is virtually no counterparty risk as the notional is not exchanged. There is no financing risk. The swap curve is smooth and not subject to QE-implied volatility. Second, short-dated implied volatility on Treasuries now trades at a premium over short-dated volatility on swaps. That used to be the other way around (because swap spreads used to be directional, amplifying a move up/down in interest rates). Now, 3-mo-into-15yr TSY vol trades about 10% over the equivalent swaption vol. Risk-adjusted Treasury yields don't look as attractive as they otherwise would be compared to credit curves and this could accelerate a possible move out of Treasuries (similar, maybe, to the move from CMT into CMS in the 80s).

Conclusions
The Fed is watching inflation expectations and is fearful that, if expectations fall, deflation could ensue. Given the low level of inflation and inflation expectations against the soft economic backdrop nearly any major shock could substantially lower inflation expectations. In such a case the Federal Reserve would very likely adopt a second-round QE program. The next round of QE would more likely consist of Treasury securities instead of mortgages. The announcement and initial rounds of Treasury purchases by the Fed would lower yields and flatten the Treasury curve. This is more due to a zero-bound constraint and already distant market expectation for rate hikes by the Fed. Further, this expectation implies that inflation expectations are stable or falling. The market is already pricing in such a scenario. The high risk here is that once the market perceives potential success from QE for the economy, the inflation expectation could rise swiftly. There is greater chance of this turn in comparison to Japan where low inflation and inflation expectations remained intact. Further, we would be wary that major Treasury holders sensing success of the QE plan would use the Fed buying program to unload.

9 August 2010

This document is being provided for the exclusive use of MARTIN ROSE (SGCIB)

IMPORTANT DISCLAIMER: The information herein is not intended to be an offer to buy or sell, or a solicitation of an offer to buy or sell, any securities and including any expression of opinion, has been obtained from or is based upon sources believed to be reliable but is not guaranteed as to accuracy or completeness although Socit Gnrale (SG) believe it to be clear, fair and not misleading. SG, and their affiliated companies in the SG Group, may from time to time deal in, profit from the trading of, hold or act as market-makers or act as advisers, brokers or bankers in relation to the securities, or derivatives thereof, of persons, firms or entities mentioned in this document or be represented on the board of such persons, firms or entities. SG is acting as a principal trader in debt securities that may be referred to in this report and may hold debt securities positions. Employees of SG, and their affiliated companies in the SG Group, or individuals connected to then, other than the authors of this report, may from time to time have a position in or be holding any of the investments or related investments mentioned in this document. Each author of this report is not permitted to trade in or hold any of the investments or related investments which are the subject of this document. SG and their affiliated companies in the SG Group are under no obligation to disclose or take account of this document when advising or dealing with or for their customers. The views of SG reflected in this document may change without notice. To the maximum extent possible at law, SG does not accept any liability whatsoever arising from the use of the material or information contained herein. This research document is not intended for use by or targeted at retail customers. Should a retail customer obtain a copy of this report they should not base their investment decisions solely on the basis of this document but must seek independent financial advice. Important notice: The circumstances in which materials provided by SG Fixed & Forex Research, SG Commodity Research, SG Convertible Research and SG Equity Derivatives Research have been produced are such (for example because of reporting or remuneration structures or the physical location of the author of the material) that it is not appropriate to characterise it as independent investment research as referred to in European MIF directive and that it should be treated as a marketing material even if it contains a research recommendation ( recommandation dinvestissement caractre promotionnel ). However, it must be made clear that all publications issued by SG will be clear, fair, and not misleading. Analyst Certification: Each author of this research report hereby certifies that (i) the views expressed in the research report accurately reflect his or her personal views about any and all of the subject securities or issuers and (ii) no part of his or her compensation was, is, or will be related, directly or indirectly, to the specific recommendations or views expressed in this report. Notice to French Investors: This publication is issued in France by or through Socit Gnrale ("SG") which is authorised by the CECEI and regulated by the AMF (Autorit des Marchs Financiers). Notice to UK investors: This publication is issued in the United Kingdom by or through Socit Gnrale ("SG") London Branch which is regulated by the Financial Services Authority ("FSA") for the conduct of its UK business. Notice To US Investors: This report is intended only for major US institutional investors pursuant to SEC Rule 15a-6. Any US person wishing to discuss this report or effect transactions in any security discussed herein should do so with or through SG Americas Securities, LLC (SGAS) 1221 Avenue of the Americas, New York, NY 10020. (212)-278-6000. THIS RESEARCH REPORT IS PRODUCED BY SOCIETE GENERALE AND NOT SGAS. Notice to Japanese Investors: This report is distributed in Japan by Socit Gnrale Securities (North Pacific) Ltd., Tokyo Branch, which is regulated by the Financial Services Agency of Japan. The products mentioned in this report may not be eligible for sale in Japan and they may not be suitable for all types of investors. Notice to Australian Investors: Socit Gnrale Australia Branch (ABN 71 092 516 286) (SG) takes responsibility for publishing this document. SG holds an AFSL no. 236651 issued under the Corporations Act 2001 (Cth) ("Act"). The information contained in this newsletter is only directed to recipients who are wholesale clients as defined under the Act. IMPORTANT DISCLOSURES: Please refer to our websites: http://www.sgresearch.socgen.com/compliance.rha http://www.sgcib.com. Copyright: The Socit Gnrale Group 2010. All rights reserved.

10

9 August 2010

This document is being provided for the exclusive use of MARTIN ROSE (SGCIB)